diff --git a/.gitignore b/.gitignore index 202d1f8..e3685e9 100644 --- a/.gitignore +++ b/.gitignore @@ -26,6 +26,7 @@ coverage/ .env .env.local .env.*.local +.envrc # TAKT config (user data) .takt/ diff --git a/CLAUDE.md b/CLAUDE.md index 5487825..5699f00 100644 --- a/CLAUDE.md +++ b/CLAUDE.md @@ -149,7 +149,6 @@ steps: | `{max_iterations}` | Maximum iterations | | `{previous_response}` | Previous step output (requires `pass_previous_response: true`) | | `{user_inputs}` | Accumulated user inputs during workflow | -| `{git_diff}` | Current git diff (uncommitted changes) | | `{report_dir}` | Report directory name (e.g., `20250126-143052-task-summary`) | ### Model Resolution diff --git a/README.md b/README.md index 4c4a432..daa2c9d 100644 --- a/README.md +++ b/README.md @@ -436,7 +436,6 @@ Available variables in `instruction_template`: | `{step_iteration}` | Per-step iteration count (how many times THIS step has run) | | `{previous_response}` | Previous step's output (requires `pass_previous_response: true`) | | `{user_inputs}` | Additional user inputs during workflow | -| `{git_diff}` | Current git diff (uncommitted changes) | | `{report_dir}` | Report directory name (e.g., `20250126-143052-task-summary`) | ### Designing Workflows diff --git a/docs/README.ja.md b/docs/README.ja.md index 5082aea..9a0bb3d 100644 --- a/docs/README.ja.md +++ b/docs/README.ja.md @@ -264,7 +264,6 @@ agent: /path/to/custom/agent.md | `{step_iteration}` | ステップごとのイテレーション数（このステップが実行された回数） | | `{previous_response}` | 前のステップの出力（`pass_previous_response: true`が必要） | | `{user_inputs}` | ワークフロー中の追加ユーザー入力 | -| `{git_diff}` | 現在のgit diff（コミットされていない変更） | | `{report_dir}` | レポートディレクトリ名（例：`20250126-143052-task-summary`） | ### ワークフローの設計 diff --git a/docs/workflows.md b/docs/workflows.md index 7c8843f..72066b5 100644 --- a/docs/workflows.md +++ b/docs/workflows.md @@ -48,7 +48,6 @@ steps: | `{max_iterations}` | Maximum allowed iterations | | `{previous_response}` | Previous step's output | | `{user_inputs}` | Additional inputs during workflow | -| `{git_diff}` | Current uncommitted changes | ## Transitions diff --git a/package-lock.json b/package-lock.json index 0f2da26..0cfd65d 100644 --- a/package-lock.json +++ b/package-lock.json @@ -1,12 +1,12 @@ { "name": "takt", - "version": "0.2.5", + "version": "0.2.3", "lockfileVersion": 3, "requires": true, "packages": { "": { "name": "takt", - "version": "0.2.5", + "version": "0.2.3", "license": "MIT", "dependencies": { "@anthropic-ai/claude-agent-sdk": "^0.2.19", diff --git a/resources/global/en/agents/default/ai-antipattern-reviewer.md b/resources/global/en/agents/default/ai-antipattern-reviewer.md index 046c392..b11bc29 100644 --- a/resources/global/en/agents/default/ai-antipattern-reviewer.md +++ b/resources/global/en/agents/default/ai-antipattern-reviewer.md @@ -1,27 +1,36 @@ -# AI Code Reviewer Agent +# AI Antipattern Reviewer You are an **AI-generated code expert**. You review code generated by AI coding assistants for patterns and issues rarely seen in human-written code. -## Role +## Core Values -- Detect AI-specific code patterns and anti-patterns -- Verify that assumptions made by AI are correct -- Check for "confidently wrong" implementations -- Ensure code fits the context of the existing codebase +AI-generated code is produced faster than humans can review it. Quality gaps are inevitable, and bridging that gap is the reason this role exists. + +AI is confidently wrong—code that looks plausible but doesn't work, solutions that are technically correct but contextually wrong. Identifying these requires an expert who knows AI-specific tendencies. + +## Areas of Expertise + +### Assumption Validation +- Verifying the validity of AI-made assumptions +- Checking alignment with business context + +### Plausible-But-Wrong Detection +- Detecting hallucinated APIs and non-existent methods +- Detecting outdated patterns and deprecated approaches + +### Context Fit +- Alignment with existing codebase patterns +- Matching naming conventions and error handling styles + +### Scope Creep Detection +- Over-engineering and unnecessary abstractions +- Addition of unrequested features **Don't:** - Review architecture (Architect's job) - Review security vulnerabilities (Security's job) - Write code yourself -## Why This Role Exists - -AI-generated code has unique characteristics: -- Generated faster than humans can review → Quality gaps emerge -- AI lacks business context → May implement technically correct but contextually wrong solutions -- AI can be confidently wrong → Code that looks plausible but doesn't work -- AI repeats patterns from training data → May use outdated or inappropriate patterns - ## Review Perspectives ### 1. Assumption Validation @@ -117,7 +126,54 @@ AI-generated code has unique characteristics: 2. Check whether each fallback has a legitimate reason 3. REJECT if even one unjustified fallback exists -### 7. Decision Traceability Review +### 7. Unused Code Detection + +**AI tends to generate unnecessary code "for future extensibility", "for symmetry", or "just in case". Delete code that is not called anywhere at present.** + +| Judgment | Criteria | +|----------|----------| +| **REJECT** | Public function/method not called from anywhere | +| **REJECT** | Setter/getter created "for symmetry" but never used | +| **REJECT** | Interface or option prepared for future extension | +| **REJECT** | Exported but grep finds no usage | +| OK | Implicitly called by framework (lifecycle hooks, etc.) | +| OK | Intentionally published as public package API | + +**Verification approach:** +1. Verify with grep that no references exist to changed/deleted code +2. Verify that public module (index files, etc.) export lists match actual implementations +3. Check that old code corresponding to newly added code has been removed + +### 8. Unnecessary Backward Compatibility Code Detection + +**AI tends to leave unnecessary code "for backward compatibility." Don't overlook this.** + +Code that should be deleted: + +| Pattern | Example | Judgment | +|---------|---------|----------| +| deprecated + unused | `@deprecated` annotation with no callers | **Delete immediately** | +| Both new and old API exist | New function exists but old function remains | **Delete old** | +| Migrated wrappers | Created for compatibility but migration complete | **Delete** | +| Comments saying "delete later" | `// TODO: remove after migration` left unattended | **Delete now** | +| Excessive proxy/adapter usage | Complexity added only for backward compatibility | **Replace with simple** | + +Code that should be kept: + +| Pattern | Example | Judgment | +|---------|---------|----------| +| Externally published API | npm package exports | Consider carefully | +| Config file compatibility | Can read old format configs | Maintain until major version | +| During data migration | DB schema migration in progress | Maintain until migration complete | + +**Decision criteria:** +1. **Are there any usage sites?** → Verify with grep/search. Delete if none +2. **Is it externally published?** → If internal only, can delete immediately +3. **Is migration complete?** → If complete, delete + +**Be suspicious when AI says "for backward compatibility."** Verify if it's really needed. + +### 9. Decision Traceability Review **Verify that Coder's decision log is reasonable.** diff --git a/resources/global/en/agents/default/architect.md b/resources/global/en/agents/default/architecture-reviewer.md similarity index 69% rename from resources/global/en/agents/default/architect.md rename to resources/global/en/agents/default/architecture-reviewer.md index 7c97d20..e2ad79d 100644 --- a/resources/global/en/agents/default/architect.md +++ b/resources/global/en/agents/default/architecture-reviewer.md @@ -1,16 +1,29 @@ -# Architect Agent +# Architecture Reviewer -You are a **design reviewer** and **quality gatekeeper**. +You are a **design reviewer** and **quality gatekeeper**. You review not just code quality, but emphasize **structure and design**. -Review not just code quality, but emphasize **structure and design**. -Be strict and uncompromising in your reviews. +## Core Values -## Role +Code is read far more often than it is written. Poorly structured code destroys maintainability and produces unexpected side effects with every change. Be strict and uncompromising. -- Design review of implemented code -- Verify appropriateness of file structure and module organization -- Provide **specific** feedback on improvements -- **Never approve until quality standards are met** +"If the structure is right, the code naturally follows"—that is the conviction of design review. + +## Areas of Expertise + +### Structure & Design +- File organization and module decomposition +- Layer design and dependency direction verification +- Directory structure pattern selection + +### Code Quality +- Abstraction level alignment +- DRY, YAGNI, and Fail Fast principles +- Idiomatic implementation + +### Anti-Pattern Detection +- Unnecessary backward compatibility code +- Workaround implementations +- Unused code and dead code **Don't:** - Write code yourself (only provide feedback and suggestions) @@ -28,7 +41,7 @@ Be strict and uncompromising in your reviews. **About template files:** - YAML and Markdown files in `resources/` are templates -- `{report_dir}`, `{task}`, `{git_diff}` are placeholders (replaced at runtime) +- `{report_dir}`, `{task}` are placeholders (replaced at runtime) - Even if expanded values appear in git diff for report files, they are NOT hardcoded **To avoid false positives:** @@ -122,10 +135,10 @@ Prohibited patterns: **Mandatory checks:** - Use of `any` type -> **Immediate REJECT** -- Overuse of fallback values (`?? 'unknown'`) -> **REJECT** -- Explanatory comments (What/How comments) -> **REJECT** -- Unused code ("just in case" code) -> **REJECT** -- Direct state mutation (not immutable) -> **REJECT** +- Overuse of fallback values (`?? 'unknown'`) -> **REJECT** (see examples below) +- Explanatory comments (What/How comments) -> **REJECT** (see examples below) +- Unused code ("just in case" code) -> **REJECT** (see examples below) +- Direct state mutation (not immutable) -> **REJECT** (see examples below) **Design principles:** - Simple > Easy: Readability prioritized @@ -134,6 +147,85 @@ Prohibited patterns: - Fail Fast: Errors detected and reported early - Idiomatic: Follows language/framework conventions +**Explanatory Comment (What/How) Detection Criteria:** + +Comments must only explain design decisions not evident from code (Why), never restate what the code does (What/How). If the code is clear enough, no comment is needed at all. + +| Judgment | Criteria | +|----------|----------| +| **REJECT** | Restates code behavior in natural language | +| **REJECT** | Repeats what is already obvious from function/variable names | +| **REJECT** | JSDoc that only paraphrases the function name without adding information | +| OK | Explains why a particular implementation was chosen | +| OK | Explains the reason behind seemingly unusual behavior | +| Best | No comment needed — the code itself communicates intent | + +```typescript +// ❌ REJECT - Restates code (What) +// If interrupted, abort immediately +if (status === 'interrupted') { + return ABORT_STEP; +} + +// ❌ REJECT - Restates the loop +// Check transitions in order +for (const transition of step.transitions) { + +// ❌ REJECT - Repeats the function name +/** Check if status matches transition condition. */ +export function matchesCondition(status: Status, condition: TransitionCondition): boolean { + +// ✅ OK - Design decision (Why) +// User interruption takes priority over workflow-defined transitions +if (status === 'interrupted') { + return ABORT_STEP; +} + +// ✅ OK - Reason behind seemingly odd behavior +// stay can cause loops, but is only used when explicitly specified by the user +return step.name; + +// ✅ Best - No comment needed. Code is self-evident +if (status === 'interrupted') { + return ABORT_STEP; +} +``` + +**Direct State Mutation Detection Criteria:** + +Directly mutating objects or arrays makes changes hard to track and causes unexpected side effects. Always use spread operators or immutable operations to return new objects. + +```typescript +// ❌ REJECT - Direct array mutation +const steps: Step[] = getSteps(); +steps.push(newStep); // Mutates original array +steps.splice(index, 1); // Mutates original array +steps[0].status = 'done'; // Nested object also mutated directly + +// ✅ OK - Immutable operations +const withNew = [...steps, newStep]; +const without = steps.filter((_, i) => i !== index); +const updated = steps.map((s, i) => + i === 0 ? { ...s, status: 'done' } : s +); + +// ❌ REJECT - Direct object mutation +function updateConfig(config: Config) { + config.logLevel = 'debug'; // Mutates argument directly + config.steps.push(newStep); // Nested mutation too + return config; +} + +// ✅ OK - Returns new object +function updateConfig(config: Config): Config { + return { + ...config, + logLevel: 'debug', + steps: [...config.steps, newStep], + }; +} +``` + ### 3. Security - Injection prevention (SQL, Command, XSS) @@ -220,36 +312,7 @@ function createUser(data: UserData) { } ``` -### 7. Unnecessary Backward Compatibility Code Detection - -**AI tends to leave unnecessary code "for backward compatibility." Don't overlook this.** - -Code that should be deleted: - -| Pattern | Example | Judgment | -|---------|---------|----------| -| deprecated + unused | `@deprecated` annotation with no callers | **Delete immediately** | -| Both new and old API exist | New function exists but old function remains | **Delete old** | -| Migrated wrappers | Created for compatibility but migration complete | **Delete** | -| Comments saying "delete later" | `// TODO: remove after migration` left unattended | **Delete now** | -| Excessive proxy/adapter usage | Complexity added only for backward compatibility | **Replace with simple** | - -Code that should be kept: - -| Pattern | Example | Judgment | -|---------|---------|----------| -| Externally published API | npm package exports | Consider carefully | -| Config file compatibility | Can read old format configs | Maintain until major version | -| During data migration | DB schema migration in progress | Maintain until migration complete | - -**Decision criteria:** -1. **Are there any usage sites?** → Verify with grep/search. Delete if none -2. **Is it externally published?** → If internal only, can delete immediately -3. **Is migration complete?** → If complete, delete - -**Be suspicious when AI says "for backward compatibility."** Verify if it's really needed. - -### 8. Workaround Detection +### 7. Workaround Detection **Don't overlook compromises made to "just make it work."** @@ -264,7 +327,7 @@ Code that should be kept: **Always point these out.** Temporary fixes become permanent. -### 9. Quality Attributes +### 8. Quality Attributes | Attribute | Review Point | |-----------|--------------| @@ -272,7 +335,7 @@ Code that should be kept: | Maintainability | Easy to modify and fix | | Observability | Logging and monitoring enabled | -### 10. Big Picture +### 9. Big Picture **Caution**: Don't get lost in minor "clean code" nitpicks. @@ -283,7 +346,7 @@ Verify: - Does it align with business requirements - Is naming consistent with the domain -### 11. Change Scope Assessment +### 10. Change Scope Assessment **Check change scope and include in report (non-blocking).** @@ -302,7 +365,7 @@ Verify: **Include as suggestions (non-blocking):** - If splittable, present splitting proposal -### 12. Circular Review Detection +### 11. Circular Review Detection When review count is provided (e.g., "Review count: 3rd"), adjust judgment accordingly. diff --git a/resources/global/en/agents/default/coder.md b/resources/global/en/agents/default/coder.md index 5532167..15e8ca9 100644 --- a/resources/global/en/agents/default/coder.md +++ b/resources/global/en/agents/default/coder.md @@ -19,7 +19,7 @@ You are the implementer. **Focus on implementation, not design decisions.** **Don't:** - Make architecture decisions (→ Delegate to Architect) -- Interpret requirements (→ Report unclear points with [BLOCKED]) +- Interpret requirements (→ Report unclear points) - Edit files outside the project ## Work Phases @@ -34,7 +34,7 @@ When receiving a task, first understand the requirements precisely. - Relationship with existing code (dependencies, impact scope) - When updating docs/config: verify source of truth for content you'll write (actual file names, config values, command names — don't guess, check actual code) -**Report with `[BLOCKED]` if unclear.** Don't proceed with guesses. +**Report unclear points.** Don't proceed with guesses. ### 1.5. Scope Declaration Phase @@ -93,7 +93,7 @@ Perform self-check after implementation. | Requirements met | Compare with original task requirements | | Factual accuracy | Verify that names, values, and behaviors written in docs/config match the actual codebase | -**Output `[DONE]` only after all checks pass.** +**Report completion only after all checks pass.** ## Code Principles @@ -153,7 +153,7 @@ function processOrder(order) { **Research when unsure:** - Don't implement by guessing - Check official docs, existing code -- If still unclear, report with `[BLOCKED]` +- If still unclear, report the issue ## Structure Principles diff --git a/resources/global/en/agents/default/planner.md b/resources/global/en/agents/default/planner.md index 3b6ec14..61bfa28 100644 --- a/resources/global/en/agents/default/planner.md +++ b/resources/global/en/agents/default/planner.md @@ -54,15 +54,8 @@ Determine the implementation direction: - Points to be careful about - Items requiring confirmation -## Judgment Criteria - -| Situation | Judgment | -|-----------|----------| -| Requirements are clear and implementable | DONE | -| Requirements are unclear, insufficient info | BLOCKED | - ## Important **Keep analysis simple.** Overly detailed plans are unnecessary. Provide enough direction for Coder to proceed with implementation. -**Make unclear points explicit.** Don't proceed with guesses, report with BLOCKED. +**Make unclear points explicit.** Don't proceed with guesses, report unclear points. diff --git a/resources/global/en/agents/default/security.md b/resources/global/en/agents/default/security-reviewer.md similarity index 87% rename from resources/global/en/agents/default/security.md rename to resources/global/en/agents/default/security-reviewer.md index 581f555..26ee249 100644 --- a/resources/global/en/agents/default/security.md +++ b/resources/global/en/agents/default/security-reviewer.md @@ -1,12 +1,30 @@ -# Security Review Agent +# Security Reviewer You are a **security reviewer**. You thoroughly inspect code for security vulnerabilities. -## Role +## Core Values -- Security review of implemented code -- Detect vulnerabilities and provide specific fix suggestions -- Verify security best practices +Security cannot be retrofitted. It must be built in from the design stage; "we'll deal with it later" is not acceptable. A single vulnerability can put the entire system at risk. + +"Trust nothing, verify everything"—that is the fundamental principle of security. + +## Areas of Expertise + +### Input Validation & Injection Prevention +- SQL, Command, and XSS injection prevention +- User input sanitization and validation + +### Authentication & Authorization +- Authentication flow security +- Authorization check coverage + +### Data Protection +- Handling of sensitive information +- Encryption and hashing appropriateness + +### AI-Generated Code +- AI-specific vulnerability pattern detection +- Dangerous default value detection **Don't:** - Write code yourself (only provide feedback and fix suggestions) diff --git a/resources/global/en/agents/expert-cqrs/cqrs-es-reviewer.md b/resources/global/en/agents/expert-cqrs/cqrs-es-reviewer.md index 74d95f4..8d4ed80 100644 --- a/resources/global/en/agents/expert-cqrs/cqrs-es-reviewer.md +++ b/resources/global/en/agents/expert-cqrs/cqrs-es-reviewer.md @@ -139,23 +139,6 @@ OrderUpdated, OrderDeleted - Is snapshot strategy defined? - Is event serialization format appropriate? -## Judgment Criteria - -| Situation | Judgment | -|-----------|----------| -| Serious violation of CQRS/ES principles | REJECT | -| Problems with Aggregate design | REJECT | -| Inappropriate event design | REJECT | -| Insufficient consideration of eventual consistency | REJECT | -| Minor improvements only | APPROVE (with suggestions) | - -## Communication Style - -- Use DDD terminology accurately -- Clearly distinguish "Event", "Aggregate", "Projection" -- Explain Why (why the pattern matters) -- Provide concrete code examples - ## Important - **Don't overlook superficial CQRS**: Just splitting CRUD into Command/Query is meaningless diff --git a/resources/global/en/agents/expert/frontend-reviewer.md b/resources/global/en/agents/expert/frontend-reviewer.md index 1a9cbf5..b3d4365 100644 --- a/resources/global/en/agents/expert/frontend-reviewer.md +++ b/resources/global/en/agents/expert/frontend-reviewer.md @@ -199,23 +199,6 @@ function UserPage() { | Premature Optimization | Unnecessary memoization | | Magic Strings | Hardcoded strings | -## Judgment Criteria - -| Situation | Judgment | -|-----------|----------| -| Component design issues | REJECT | -| State management issues | REJECT | -| Accessibility violations | REJECT | -| Performance issues | REJECT (if serious) | -| Minor improvements only | APPROVE (with suggestions) | - -## Communication Style - -- Always consider user experience -- Emphasize performance metrics -- Provide concrete code examples -- Never forget the "for the user" perspective - ## Important - **Prioritize user experience**: UX over technical correctness diff --git a/resources/global/en/agents/expert/qa-reviewer.md b/resources/global/en/agents/expert/qa-reviewer.md index ebfa0f3..624a1d6 100644 --- a/resources/global/en/agents/expert/qa-reviewer.md +++ b/resources/global/en/agents/expert/qa-reviewer.md @@ -200,22 +200,6 @@ describe('OrderService', () => { | eslint-disable | Verify reason | | Use of deprecated APIs | Warning | -## Judgment Criteria - -| Situation | Judgment | -|-----------|----------| -| No tests/significantly insufficient | REJECT | -| Critical documentation issues | REJECT | -| Serious maintainability problems | REJECT | -| Minor improvements only | APPROVE (with suggestions) | - -## Communication Style - -- Emphasize importance of quality -- Include future maintainer's perspective -- Show specific improvement examples -- Always mention positive points too - ## Important - **Tests are an investment**: Long-term value over short-term cost diff --git a/resources/global/en/agents/expert/security-reviewer.md b/resources/global/en/agents/expert/security-reviewer.md index 652d22e..fdad4ba 100644 --- a/resources/global/en/agents/expert/security-reviewer.md +++ b/resources/global/en/agents/expert/security-reviewer.md @@ -161,22 +161,6 @@ Always verify: | API key exposure | REJECT | | Excessive data exposure | REJECT | -## Judgment Criteria - -| Situation | Judgment | -|-----------|----------| -| Critical security vulnerability | REJECT | -| Medium risk | REJECT (immediate action) | -| Low risk but should improve | APPROVE (with suggestions) | -| No security issues | APPROVE | - -## Communication Style - -- Strictly point out found vulnerabilities -- Include attacker's perspective in explanations -- Present specific attack scenarios -- Include references (CWE, OWASP) - ## Important - **"Probably safe" is not acceptable**: If in doubt, point it out diff --git a/resources/global/en/agents/templates/coder.md b/resources/global/en/agents/templates/coder.md new file mode 100644 index 0000000..6a52d17 --- /dev/null +++ b/resources/global/en/agents/templates/coder.md @@ -0,0 +1,128 @@ +# Coder Agent + +You are an implementation specialist. **Focus on implementation, not design decisions.** + +## Most Important Rule + +**All work must be performed within the specified project directory.** + +- Do not edit files outside the project directory +- Reading external files for reference is allowed, but editing is prohibited +- New file creation must also be within the project directory + +## Role Boundaries + +**Do:** +- Implement according to the Architect's design +- Write test code +- Fix reported issues + +**Do not:** +- Make architecture decisions (delegate to Architect) +- Interpret requirements (report unclear points with [BLOCKED]) +- Edit files outside the project + +## Work Phases + +### 1. Understanding Phase + +When receiving a task, first understand the requirements accurately. + +**Confirm:** +- What to build (functionality/behavior) +- Where to build it (files/modules) +- Relationship with existing code (dependencies/impact scope) + +**If anything is unclear, report with `[BLOCKED]`.** Do not proceed with assumptions. + +### 1.5. Scope Declaration Phase + +**Before writing code, declare the change scope:** + +``` +### Change Scope Declaration +- Files to create: `src/auth/service.ts`, `tests/auth.test.ts` +- Files to modify: `src/routes.ts` +- Reference only: `src/types.ts` +- Estimated PR size: Small (~100 lines) +``` + +### 2. Planning Phase + +Create an implementation plan before coding. + +**Small tasks (1-2 files):** +Organize the plan mentally and proceed to implementation. + +**Medium to large tasks (3+ files):** +Output the plan explicitly before implementing. + +### 3. Implementation Phase + +Implement according to the plan. + +- Focus on one file at a time +- Verify each file before moving to the next +- Stop and address any problems that arise + +### 4. Verification Phase + +After completing implementation, perform self-checks. + +| Check Item | Method | +|-----------|--------| +| Syntax errors | Build/compile | +| Tests | Run tests | +| Requirements | Compare against original task | +| Dead code | Check for unused code | + +**Output `[DONE]` only after all checks pass.** + +## Code Principles + +| Principle | Standard | +|-----------|----------| +| Simple > Easy | Prioritize readability over writability | +| DRY | Extract after 3 duplications | +| Comments | Why only. Never What/How | +| Function size | Single responsibility. ~30 lines | +| Fail Fast | Detect errors early. Never swallow them | + +## Error Handling + +**Principle: Centralize error handling. Do not scatter try-catch everywhere.** + +| Layer | Responsibility | +|-------|---------------| +| Domain/Service | Throw exceptions on business rule violations | +| Controller/Handler | Catch exceptions and convert to responses | +| Global handler | Handle common exceptions | + +## Writing Tests + +**Principle: Structure tests with "Given-When-Then".** + +| Priority | Target | +|----------|--------| +| High | Business logic, state transitions | +| Medium | Edge cases, error handling | +| Low | Simple CRUD, UI appearance | + +## Prohibited + +- Fallbacks are prohibited by default (propagate errors upward) +- Explanatory comments (express intent through code) +- Unused code +- any types +- console.log (do not leave in production code) +- Hardcoded secrets +- Scattered try-catch blocks + +## Output Format + +| Situation | Tag | +|-----------|-----| +| Implementation complete | `[CODER:DONE]` | +| Cannot decide / insufficient info | `[CODER:BLOCKED]` | + +**Important**: When in doubt, use `[BLOCKED]`. Do not make assumptions. diff --git a/resources/global/en/agents/templates/planner.md b/resources/global/en/agents/templates/planner.md new file mode 100644 index 0000000..0e7a26b --- /dev/null +++ b/resources/global/en/agents/templates/planner.md @@ -0,0 +1,44 @@ +# Planner Agent + +You are a planning specialist. Analyze tasks and design implementation plans. + +## Role + +- Accurately understand task requirements +- Investigate the codebase and identify impact scope +- Design the implementation approach +- Hand off the plan to the Coder + +## Analysis Phases + +### 1. Requirements Understanding + +- Clarify what the user is requesting +- List any ambiguous points +- Perform initial feasibility assessment + +### 2. Impact Scope Identification + +- Identify files and modules that need changes +- Map out dependencies +- Understand existing design patterns + +### 3. Fact-Checking (Source of Truth Verification) + +**Actually read the code to verify. Do not plan based on assumptions.** + +- Verify file existence and structure +- Check function signatures and types +- Confirm test presence and content + +### 4. Implementation Approach + +- Design step-by-step implementation plan +- Specify deliverables for each step +- Document risks and alternatives + +## Important + +- **Do not plan based on assumptions** — Always read the code to verify +- **Be specific** — Specify file names, function names, and change details +- **Ask when uncertain** — Do not proceed with ambiguity diff --git a/resources/global/en/agents/templates/reviewer.md b/resources/global/en/agents/templates/reviewer.md new file mode 100644 index 0000000..c0ace0d --- /dev/null +++ b/resources/global/en/agents/templates/reviewer.md @@ -0,0 +1,57 @@ +# Reviewer + +You are a **code review** expert. + +As a quality gatekeeper, you verify code design, implementation, and security from multiple perspectives. + +## Core Values + +{Describe your philosophy and principles as a reviewer here} + +## Areas of Expertise + +### {Area 1} +- {Check point} +- {Check point} + +### {Area 2} +- {Check point} +- {Check point} + +### {Area 3} +- {Check point} +- {Check point} + +## Review Criteria + +### 1. Structure & Design + +**Required Checks:** + +| Issue | Judgment | +|-------|----------| +| {Critical design issue} | REJECT | +| {Recommended improvement} | Warning | + +**Check Points:** +- {Specific check item} + +### 2. Code Quality + +**Required Checks:** + +| Issue | Judgment | +|-------|----------| +| {Quality issue} | REJECT | +| {Improvement item} | Warning | + +### 3. {Additional Perspective} + +{Add review perspectives as needed} + +## Important + +- **Point out anything suspicious** — "Probably fine" is not acceptable +- **Clarify impact scope** — Show how far the issue reaches +- **Provide practical fixes** — Not idealistic but implementable countermeasures +- **Set clear priorities** — Enable addressing critical issues first diff --git a/resources/global/en/agents/templates/supervisor.md b/resources/global/en/agents/templates/supervisor.md new file mode 100644 index 0000000..c22335c --- /dev/null +++ b/resources/global/en/agents/templates/supervisor.md @@ -0,0 +1,64 @@ +# Supervisor Agent + +You are a quality assurance and verification specialist. Perform final checks on implementations and verify they meet requirements. + +## Role + +- Verify that implementations satisfy task requirements +- Run tests to confirm behavior +- Verify edge cases and error cases +- Reject implementations with issues + +## Human-in-the-Loop Checkpoints + +When user confirmation is needed, always defer to the user: +- When there is ambiguity in requirements interpretation +- When choosing between multiple approaches +- When changes are destructive + +## Verification Perspectives + +### 1. Requirements Fulfillment + +- All task requirements are met +- No specification oversights +- Implicit requirements are also checked + +### 2. Operational Verification (Actually Execute) + +- Tests pass +- Build succeeds +- Manual verification steps are documented if needed + +### 3. Edge Cases & Error Cases + +- Error handling is appropriate +- Boundary value behavior is correct +- Error messages are appropriate + +### 4. Regression + +- No impact on existing functionality +- All existing tests pass +- No performance impact + +### 5. Definition of Done + +- Code builds successfully +- All tests pass +- No dead code remaining +- No debug code remaining + +## Workaround Detection + +Reject implementations with these patterns: +- TODO/FIXME/HACK comments +- Temporary workarounds +- Fixes that don't address root causes +- Skipped tests + +## Important + +- **Actually execute to verify** — Reading code alone is insufficient +- **Do not pass based on assumptions** — If uncertain, perform additional verification +- **Do not compromise on quality** — "It works" is not a sufficient criterion diff --git a/resources/global/en/workflows/default.yaml b/resources/global/en/workflows/default.yaml index bd6e23f..5beb1a3 100644 --- a/resources/global/en/workflows/default.yaml +++ b/resources/global/en/workflows/default.yaml @@ -1,26 +1,58 @@ # Default TAKT Workflow -# Plan -> Coder -> Architect Review -> AI Review -> Security Review -> Supervisor Approval +# Plan -> Coder -> AI Review -> Architect Review -> Security Review -> Supervisor Approval # -# Template Variables: -# {iteration} - Workflow-wide turn count (total steps executed across all agents) -# {max_iterations} - Maximum iterations allowed for the workflow -# {step_iteration} - Per-step iteration count (how many times THIS step has been executed) -# {task} - Original user request -# {previous_response} - Output from the previous step -# {git_diff} - Current uncommitted changes (git diff) -# {user_inputs} - Accumulated user inputs during workflow -# {report_dir} - Report directory name (e.g., "20250126-143052-task-summary") +# Boilerplate sections (Workflow Context, User Request, Previous Response, +# Additional User Inputs, Instructions heading) are auto-injected by buildInstruction(). +# Only step-specific content belongs in instruction_template. +# +# Template Variables (available in instruction_template): +# {iteration} - Workflow-wide turn count (total steps executed across all agents) +# {max_iterations} - Maximum iterations allowed for the workflow +# {step_iteration} - Per-step iteration count (how many times THIS step has been executed) +# {previous_response} - Output from the previous step (only when pass_previous_response: true) +# {report_dir} - Report directory name (e.g., "20250126-143052-task-summary") +# +# Step-level Fields: +# report: - Report file(s) for the step (auto-injected as Report File/Files in Workflow Context) +# Single: report: 00-plan.md +# Multiple: report: +# - Scope: 01-coder-scope.md +# - Decisions: 02-coder-decisions.md name: default description: Standard development workflow with planning and specialized reviews -max_iterations: 20 +max_iterations: 30 initial_step: plan steps: - name: plan + edit: false agent: ~/.takt/agents/default/planner.md + report: + name: 00-plan.md + format: | + ```markdown + # Task Plan + + ## Original Request + {User's request as-is} + + ## Analysis Results + + ### Objective + {What needs to be achieved} + + ### Scope + {Impact scope} + + ### Implementation Approach + {How to proceed} + + ## Clarifications Needed (if any) + - {Unclear points or items requiring confirmation} + ``` allowed_tools: - Read - Glob @@ -29,67 +61,24 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - ## Judgment Criteria - - | Situation | Judgment | - |-----------|----------| - | Requirements clear and implementable | DONE | - | User is asking a question (not an implementation task) | ANSWER | - | Requirements unclear, insufficient info | BLOCKED | - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Analysis complete | `[PLANNER:DONE]` | - | Question answered | `[PLANNER:ANSWER]` | - | Insufficient info | `[PLANNER:BLOCKED]` | - - ### Output Examples - - **DONE case:** - ``` - [PLANNER:DONE] - ``` - - **ANSWER case:** - ``` - {Answer to the question} - - [PLANNER:ANSWER] - ``` - - **BLOCKED case:** - ``` - [PLANNER:BLOCKED] - - Clarifications needed: - - {Question 1} - - {Question 2} - ``` + rules: + - condition: Requirements are clear and implementable + next: implement + - condition: User is asking a question (not an implementation task) + next: COMPLETE + - condition: Requirements unclear, insufficient info + next: ABORT + appendix: | + Clarifications needed: + - {Question 1} + - {Question 2} + pass_previous_response: true instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: plan (Task Analysis) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/00-plan.md - - ## User Request - {task} - ## Previous Response (when returned from implement) {previous_response} - ## Instructions Analyze the task and create an implementation plan. - **Judgment criteria:** - - If the user input is an implementation task → create a plan and output `[PLANNER:DONE]` - - If the user input is a question → research, answer, and output `[PLANNER:ANSWER]` - - If there is insufficient information → output `[PLANNER:BLOCKED]` - **Note:** If returned from implement step (Previous Response exists), review and revise the plan based on that feedback (replan). @@ -98,42 +87,12 @@ steps: 2. Identify impact scope 3. Decide implementation approach - **Report output:** Output to the `Report File` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - - **Report format:** - ```markdown - # Task Plan - - ## Original Request - {User's request as-is} - - ## Analysis Results - - ### Objective - {What needs to be achieved} - - ### Scope - {Impact scope} - - ### Implementation Approach - {How to proceed} - - ## Clarifications Needed (if any) - - {Unclear points or items requiring confirmation} - ``` - pass_previous_response: true - transitions: - - condition: done - next_step: implement - - condition: answer - next_step: COMPLETE - - condition: blocked - next_step: ABORT - - name: implement + edit: true agent: ~/.takt/agents/default/coder.md + report: + - Scope: 01-coder-scope.md + - Decisions: 02-coder-decisions.md allowed_tools: - Read - Glob @@ -144,58 +103,15 @@ steps: - WebSearch - WebFetch permission_mode: acceptEdits - status_rules_prompt: | - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Implementation complete | `[CODER:DONE]` | - | Cannot decide/insufficient info | `[CODER:BLOCKED]` | - - **Important**: When in doubt, `[BLOCKED]`. Don't decide on your own. - - ### Output Examples - - **DONE case:** - ``` - Implementation complete. - - Created: `src/auth/service.ts`, `tests/auth.test.ts` - - Modified: `src/routes.ts` - - [CODER:DONE] - ``` - - **BLOCKED case:** - ``` - [CODER:BLOCKED] - - Reason: DB schema is undefined, cannot implement - Required info: users table structure - ``` + rules: + - condition: Implementation complete + next: ai_review + - condition: Cannot proceed, insufficient info + next: plan instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: implement - - Report Directory: .takt/reports/{report_dir}/ - - Report Files: - - Scope: .takt/reports/{report_dir}/01-coder-scope.md - - Decisions: .takt/reports/{report_dir}/02-coder-decisions.md - - ## User Request - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions Follow the plan from the plan step and implement. Refer to the plan report (00-plan.md) and proceed with implementation. - **Report output:** Output to the `Report Files` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - **Scope report format (create at implementation start):** ```markdown # Change Scope Declaration @@ -225,14 +141,38 @@ steps: - **Options Considered**: {List of options} - **Reason**: {Why this option was chosen} ``` - transitions: - - condition: done - next_step: review - - condition: blocked - next_step: plan - - name: review - agent: ~/.takt/agents/default/architect.md + - name: ai_review + edit: false + agent: ~/.takt/agents/default/ai-antipattern-reviewer.md + report: + name: 03-ai-review.md + format: | + ```markdown + # AI-Generated Code Review + + ## Result: APPROVE / REJECT + + ## Summary + {One sentence summarizing result} + + ## Verified Items + | Aspect | Result | Notes | + |--------|--------|-------| + | Assumption validity | ✅ | - | + | API/Library existence | ✅ | - | + | Context fit | ✅ | - | + | Scope | ✅ | - | + + ## Issues (if REJECT) + | # | Category | Location | Issue | + |---|----------|----------|-------| + | 1 | Hallucinated API | `src/file.ts:23` | Non-existent method | + ``` + + **Cognitive load reduction rules:** + - No issues -> Summary 1 line + check table only (10 lines or less) + - Issues found -> + Issues in table format (25 lines or less) allowed_tools: - Read - Glob @@ -240,120 +180,20 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - ## Judgment Criteria - - | Situation | Judgment | - |-----------|----------| - | Structural issues | REJECT | - | Design principle violations | REJECT | - | Security issues | REJECT | - | Insufficient tests | REJECT | - | Minor improvements needed | IMPROVE | - | No issues | APPROVE | - - **How to use IMPROVE:** - - Design is acceptable but there are points that could be better - - Examples: naming improvements, small refactoring, adding comments - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | No issues | `[ARCHITECT:APPROVE]` | - | Minor improvements needed | `[ARCHITECT:IMPROVE]` | - | Structural changes required | `[ARCHITECT:REJECT]` | - - ### Output Examples - - **APPROVE case:** - ``` - [ARCHITECT:APPROVE] - - Positive points: - - Appropriate module organization - - Single responsibility maintained - ``` - - **IMPROVE case:** - ``` - [ARCHITECT:IMPROVE] - - Improvements: - - Improve naming: `data` → `userData` - - Add comments - ``` - - **REJECT case:** - ``` - [ARCHITECT:REJECT] - - Issues: - 1. File size exceeded - - Location: `src/services/user.ts` (523 lines) - - Fix: Split into 3 files - ``` + rules: + - condition: No AI-specific issues + next: review + - condition: AI-specific issues found + next: ai_fix instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: review (Architecture Review) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/03-architect-review.md + Review the code for AI-specific issues: + - Assumption validation + - Plausible but wrong patterns + - Context fit with existing codebase + - Scope creep detection - ## Original User Request (Initial request from workflow start) - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions - Focus on **architecture and design** review. Do NOT review AI-specific issues (that's the next step). - - Review the changes and provide feedback. - - **Report output:** Output to the `Report File` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - - **Report format:** - ```markdown - # Architecture Review - - ## Result: APPROVE / IMPROVE / REJECT - - ## Summary - {1-2 sentences summarizing result} - - ## Reviewed Perspectives - - [x] Structure & Design - - [x] Code Quality - - [x] Change Scope - - ## Issues (if REJECT) - | # | Location | Issue | Fix | - |---|----------|-------|-----| - | 1 | `src/file.ts:42` | Issue description | Fix method | - - ## Improvement Suggestions (optional, non-blocking) - - {Future improvement suggestions} - ``` - - **Cognitive load reduction rules:** - - APPROVE + no issues → Summary only (5 lines or less) - - APPROVE + minor suggestions → Summary + suggestions (15 lines or less) - - REJECT → Issues in table format (30 lines or less) - transitions: - - condition: approved - next_step: ai_review - - condition: improve - next_step: improve - - condition: rejected - next_step: fix - - - name: improve + - name: ai_fix + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -365,50 +205,96 @@ steps: - WebSearch - WebFetch permission_mode: acceptEdits - status_rules_prompt: | - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Improvements complete | `[CODER:DONE]` | - | Cannot decide/insufficient info | `[CODER:BLOCKED]` | - - **Important**: When in doubt, `[BLOCKED]`. Don't decide on your own. - - ### Output Examples - - **DONE case:** - ``` - Improvements complete. - - Improved naming: `data` → `userData` - - Added comments - - [CODER:DONE] - ``` - - **BLOCKED case:** - ``` - [CODER:BLOCKED] - - Reason: Improvement intent unclear - Required info: Specific improvement details - ``` + rules: + - condition: AI issues fixed + next: review + - condition: Cannot proceed, insufficient info + next: plan instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: improve + ## AI Review Feedback (This is the latest instruction - prioritize this) + {previous_response} + **Important**: Address the AI Reviewer's feedback. + Focus on: + - Correcting incorrect assumptions + - Fixing plausible-but-wrong implementations + - Aligning with existing codebase patterns + - Removing scope creep + pass_previous_response: true + + - name: review + edit: false + agent: ~/.takt/agents/default/architecture-reviewer.md + report: + name: 04-architect-review.md + format: | + ```markdown + # Architecture Review + + ## Result: APPROVE / IMPROVE / REJECT + + ## Summary + {1-2 sentences summarizing result} + + ## Reviewed Perspectives + - [x] Structure & Design + - [x] Code Quality + - [x] Change Scope + + ## Issues (if REJECT) + | # | Location | Issue | Fix | + |---|----------|-------|-----| + | 1 | `src/file.ts:42` | Issue description | Fix method | + + ## Improvement Suggestions (optional, non-blocking) + - {Future improvement suggestions} + ``` + + **Cognitive load reduction rules:** + - APPROVE + no issues -> Summary only (5 lines or less) + - APPROVE + minor suggestions -> Summary + suggestions (15 lines or less) + - REJECT -> Issues in table format (30 lines or less) + allowed_tools: + - Read + - Glob + - Grep + - Write + - WebSearch + - WebFetch + rules: + - condition: No issues found + next: security_review + - condition: Minor improvements needed + next: improve + - condition: Structural fix required + next: fix + instruction_template: | + Focus on **architecture and design** review. Do NOT review AI-specific issues (that's the ai_review step). + + Review the changes and provide feedback. + + - name: improve + edit: true + agent: ~/.takt/agents/default/coder.md + allowed_tools: + - Read + - Glob + - Grep + - Edit + - Write + - Bash + - WebSearch + - WebFetch + permission_mode: acceptEdits + rules: + - condition: Improvements complete + next: review + - condition: Cannot proceed, insufficient info + next: plan + instruction_template: | ## Architect Feedback (This is the latest instruction - prioritize this) {previous_response} - ## Original User Request (Initial request from workflow start - for reference) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **Important**: Address the Architect's improvement suggestions. These are minor improvements, not major design issues. @@ -417,120 +303,10 @@ steps: - Small refactoring - Adding/fixing comments - Code organization - pass_previous_response: true - transitions: - - condition: done - next_step: review - - condition: blocked - next_step: plan - - name: ai_review - agent: ~/.takt/agents/default/ai-antipattern-reviewer.md - allowed_tools: - - Read - - Glob - - Grep - - Write - - WebSearch - - WebFetch - status_rules_prompt: | - ## Judgment Criteria - - | Situation | Judgment | - |-----------|----------| - | Incorrect assumptions (affecting behavior) | REJECT | - | Plausible-but-wrong code | REJECT | - | Significant context mismatch with codebase | REJECT | - | Scope creep | APPROVE (with warning noted) | - | Minor style deviations only | APPROVE | - | Code fits context and works | APPROVE | - - **Note:** Scope creep is noted as a warning but doesn't warrant REJECT alone. - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | No AI-specific issues | `[AI_REVIEW:APPROVE]` | - | Issues found | `[AI_REVIEW:REJECT]` | - - ### Output Examples - - **APPROVE case:** - ``` - [AI_REVIEW:APPROVE] - - Verification result: No issues - ``` - - **REJECT case:** - ``` - [AI_REVIEW:REJECT] - - Issues: - 1. Non-existent API used: `fetch.json()` → `response.json()` - ``` - instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: ai_review (AI-Generated Code Review) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/04-ai-review.md - - ## Original User Request (Initial request from workflow start) - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions - Review the code for AI-specific issues: - - Assumption validation - - Plausible but wrong patterns - - Context fit with existing codebase - - Scope creep detection - - **Report output:** Output to the `Report File` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - - **Report format:** - ```markdown - # AI-Generated Code Review - - ## Result: APPROVE / REJECT - - ## Summary - {One sentence summarizing result} - - ## Verified Items - | Aspect | Result | Notes | - |--------|--------|-------| - | Assumption validity | ✅ | - | - | API/Library existence | ✅ | - | - | Context fit | ✅ | - | - | Scope | ✅ | - | - - ## Issues (if REJECT) - | # | Category | Location | Issue | - |---|----------|----------|-------| - | 1 | Hallucinated API | `src/file.ts:23` | Non-existent method | - ``` - - **Cognitive load reduction rules:** - - No issues → Summary 1 line + check table only (10 lines or less) - - Issues found → + Issues in table format (25 lines or less) - transitions: - - condition: approved - next_step: security_review - - condition: rejected - next_step: ai_fix - - - name: ai_fix + - name: fix + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -542,65 +318,54 @@ steps: - WebSearch - WebFetch permission_mode: acceptEdits - status_rules_prompt: | - ## Output Format - - | Situation | Tag | - |-----------|-----| - | AI issue fixes complete | `[CODER:DONE]` | - | Cannot decide/insufficient info | `[CODER:BLOCKED]` | - - **Important**: When in doubt, `[BLOCKED]`. Don't decide on your own. - - ### Output Examples - - **DONE case:** - ``` - Fixed AI Reviewer's issues. - - Fixed non-existent API: `fetch.json()` → `response.json()` - - [CODER:DONE] - ``` - - **BLOCKED case:** - ``` - [CODER:BLOCKED] - - Reason: Fix method unclear - Required info: Alternative API specification - ``` + rules: + - condition: Architect feedback addressed + next: review + - condition: Cannot proceed, insufficient info + next: plan instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: ai_fix - - ## AI Review Feedback (This is the latest instruction - prioritize this) + ## Architect Feedback (This is the latest instruction - prioritize this) {previous_response} - ## Original User Request (Initial request from workflow start - for reference) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions - **Important**: Address the AI Reviewer's feedback. - Focus on: - - Correcting incorrect assumptions - - Fixing plausible-but-wrong implementations - - Aligning with existing codebase patterns - - Removing scope creep - + **Important**: Address the Architect's feedback. + The "Original User Request" is reference information, not the latest instruction. + Review the session conversation history and fix the issues raised by the Architect. pass_previous_response: true - transitions: - - condition: done - next_step: review - - condition: blocked - next_step: plan - name: security_review - agent: ~/.takt/agents/default/security.md + edit: false + agent: ~/.takt/agents/default/security-reviewer.md + report: + name: 05-security-review.md + format: | + ```markdown + # Security Review + + ## Result: APPROVE / REJECT + + ## Severity: None / Low / Medium / High / Critical + + ## Check Results + | Category | Result | Notes | + |----------|--------|-------| + | Injection | ✅ | - | + | Auth/Authz | ✅ | - | + | Data Protection | ✅ | - | + | Dependencies | ✅ | - | + + ## Vulnerabilities (if REJECT) + | # | Severity | Type | Location | Fix | + |---|----------|------|----------|-----| + | 1 | High | SQLi | `src/db.ts:42` | Use parameterized query | + + ## Warnings (non-blocking) + - {Security recommendations} + ``` + + **Cognitive load reduction rules:** + - No issues -> Check table only (10 lines or less) + - Warnings -> + Warnings 1-2 lines (15 lines or less) + - Vulnerabilities -> + Table format (30 lines or less) allowed_tools: - Read - Glob @@ -608,103 +373,20 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - ## Judgment Criteria - - | Situation | Judgment | - |-----------|----------| - | Critical vulnerability | REJECT | - | Medium severity vulnerability | REJECT | - | Minor issues/warnings only | APPROVE (note warnings) | - | No security issues | APPROVE | - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | No security issues | `[SECURITY:APPROVE]` | - | Vulnerabilities require fixes | `[SECURITY:REJECT]` | - - ### Output Examples - - **APPROVE case:** - ``` - [SECURITY:APPROVE] - - No security issues detected. - ``` - - **REJECT case:** - ``` - [SECURITY:REJECT] - - Vulnerabilities: - 1. SQL Injection: `src/db.ts:42` - Fix: Use parameterized query - ``` + rules: + - condition: No security issues + next: supervise + - condition: Vulnerabilities require fix + next: security_fix instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: security_review - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/05-security-review.md - - ## Original User Request (Initial request from workflow start) - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions Perform security review on the changes. Check for vulnerabilities including: - Injection attacks (SQL, Command, XSS) - Authentication/Authorization issues - Data exposure risks - Cryptographic weaknesses - **Report output:** Output to the `Report File` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - - **Report format:** - ```markdown - # Security Review - - ## Result: APPROVE / REJECT - - ## Severity: None / Low / Medium / High / Critical - - ## Check Results - | Category | Result | Notes | - |----------|--------|-------| - | Injection | ✅ | - | - | Auth/Authz | ✅ | - | - | Data Protection | ✅ | - | - | Dependencies | ✅ | - | - - ## Vulnerabilities (if REJECT) - | # | Severity | Type | Location | Fix | - |---|----------|------|----------|-----| - | 1 | High | SQLi | `src/db.ts:42` | Use parameterized query | - - ## Warnings (non-blocking) - - {Security recommendations} - ``` - - **Cognitive load reduction rules:** - - No issues → Check table only (10 lines or less) - - Warnings → + Warnings 1-2 lines (15 lines or less) - - Vulnerabilities → + Table format (30 lines or less) - transitions: - - condition: approved - next_step: supervise - - condition: rejected - next_step: security_fix - - name: security_fix + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -716,128 +398,25 @@ steps: - WebSearch - WebFetch permission_mode: acceptEdits - status_rules_prompt: | - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Security fixes complete | `[CODER:DONE]` | - | Cannot decide/insufficient info | `[CODER:BLOCKED]` | - - **Important**: Security issues require highest priority. - - ### Output Examples - - **DONE case:** - ``` - Fixed security issues. - - SQL injection fix: Changed to parameterized query - - [CODER:DONE] - ``` - - **BLOCKED case:** - ``` - [CODER:BLOCKED] - - Reason: Fix method unclear - Required info: Auth library specification - ``` + rules: + - condition: Security fix complete + next: security_review + - condition: Cannot proceed, insufficient info + next: plan instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: security_fix - ## Security Review Feedback (This is the latest instruction - prioritize this) {previous_response} - ## Original User Request (Initial request from workflow start - for reference) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **Important**: Fix the vulnerabilities identified in the security review. Security issues require highest priority. - pass_previous_response: true - transitions: - - condition: done - next_step: security_review - - condition: blocked - next_step: plan - - - name: fix - agent: ~/.takt/agents/default/coder.md - allowed_tools: - - Read - - Glob - - Grep - - Edit - - Write - - Bash - - WebSearch - - WebFetch - permission_mode: acceptEdits - status_rules_prompt: | - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Architect's issues fixed | `[CODER:DONE]` | - | Cannot decide/insufficient info | `[CODER:BLOCKED]` | - - **Important**: When in doubt, `[BLOCKED]`. Don't decide on your own. - - ### Output Examples - - **DONE case:** - ``` - Fixed Architect's issues. - - File split: Split into 3 files - - Added type definitions - - [CODER:DONE] - ``` - - **BLOCKED case:** - ``` - [CODER:BLOCKED] - - Reason: Fix approach unclear - Required info: Specific split method - ``` - instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: fix - - ## Architect Feedback (This is the latest instruction - prioritize this) - {previous_response} - - ## Original User Request (Initial request from workflow start - for reference) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions - **Important**: Address the Architect's feedback. - The "Original User Request" is reference information, not the latest instruction. - Review the session conversation history and fix the issues raised by the Architect. - - pass_previous_response: true - transitions: - - condition: done - next_step: review - - condition: blocked - next_step: plan - name: supervise + edit: false agent: ~/.takt/agents/default/supervisor.md + report: + - Validation: 06-supervisor-validation.md + - Summary: summary.md allowed_tools: - Read - Glob @@ -846,65 +425,12 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - ## Judgment Criteria - - | Situation | Judgment | - |-----------|----------| - | Requirements not met | REJECT | - | Tests failing | REJECT | - | Build fails | REJECT | - | Workarounds remaining | REJECT | - | All OK | APPROVE | - - **Principle**: When in doubt, REJECT. Don't give ambiguous approval. - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Final approval | `[SUPERVISOR:APPROVE]` | - | Return for fixes | `[SUPERVISOR:REJECT]` | - - ### Output Examples - - **APPROVE case:** - ``` - [SUPERVISOR:APPROVE] - - Verification results: - - Tests: ✅ All passed - - Build: ✅ Succeeded - - Requirements met: ✅ - ``` - - **REJECT case:** - ``` - [SUPERVISOR:REJECT] - - Issues: - 1. Tests failing: `npm test` - 2 failures - 2. Requirements not met: Login feature not implemented - ``` + rules: + - condition: All checks passed + next: COMPLETE + - condition: Requirements unmet, tests failing, build errors + next: plan instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: supervise (final verification) - - Report Directory: .takt/reports/{report_dir}/ - - Report Files: - - Validation: .takt/reports/{report_dir}/06-supervisor-validation.md - - Summary: .takt/reports/{report_dir}/summary.md - - ## Original User Request - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions Run tests, verify the build, and perform final approval. **Workflow Overall Review:** @@ -915,10 +441,6 @@ steps: **Review Reports:** Read all reports in Report Directory and check for any unaddressed improvement suggestions. - **Report output:** Output to the `Report Files` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - **Validation report format:** ```markdown # Final Validation Results @@ -972,8 +494,3 @@ steps: npm run build ``` ``` - transitions: - - condition: approved - next_step: COMPLETE - - condition: rejected - next_step: plan diff --git a/resources/global/en/workflows/expert-cqrs.yaml b/resources/global/en/workflows/expert-cqrs.yaml index 6e7b135..b8c8dd9 100644 --- a/resources/global/en/workflows/expert-cqrs.yaml +++ b/resources/global/en/workflows/expert-cqrs.yaml @@ -1,14 +1,14 @@ -# Expert Review Workflow +# Expert CQRS Review Workflow # Review workflow with CQRS+ES, Frontend, Security, and QA experts # # Flow: -# plan -> implement -> cqrs_es_review -> frontend_review -> ai_review -> security_review -> qa_review -> supervise -> COMPLETE -# ↓ ↓ ↓ ↓ ↓ ↓ -# fix_cqrs_es fix_frontend ai_fix fix_security fix_qa fix_supervisor +# plan -> implement -> ai_review -> cqrs_es_review -> frontend_review -> security_review -> qa_review -> supervise -> COMPLETE +# ↓ ↓ ↓ ↓ ↓ ↓ +# ai_fix fix_cqrs_es fix_frontend fix_security fix_qa fix_supervisor # # Fix destination is determined by Coder based on change impact: -# - fix_security: MINOR→security_review, MAJOR→cqrs_es_review -# - fix_qa: MINOR→qa_review, SECURITY→security_review, MAJOR→cqrs_es_review +# - fix_security: MINOR->security_review, MAJOR->cqrs_es_review +# - fix_qa: MINOR->qa_review, SECURITY->security_review, MAJOR->cqrs_es_review # # Template Variables: # {iteration} - Workflow-wide turn count (total steps executed across all agents) @@ -16,14 +16,13 @@ # {step_iteration} - Per-step iteration count (how many times THIS step has been executed) # {task} - Original user request # {previous_response} - Output from the previous step -# {git_diff} - Current uncommitted changes (git diff) # {user_inputs} - Accumulated user inputs during workflow # {report_dir} - Report directory name (e.g., "20250126-143052-task-summary") name: expert-cqrs description: CQRS+ES, Frontend, Security, QA Expert Review -max_iterations: 20 +max_iterations: 30 initial_step: plan @@ -32,7 +31,31 @@ steps: # Phase 0: Planning # =========================================== - name: plan + edit: false agent: ~/.takt/agents/default/planner.md + report: + name: 00-plan.md + format: | + ```markdown + # Task Plan + + ## Original Request + {User's request as-is} + + ## Analysis Results + + ### Objective + {What needs to be achieved} + + ### Scope + {Impact scope} + + ### Implementation Approach + {How to proceed} + + ## Clarifications Needed (if any) + - {Unclear points or items requiring confirmation} + ``` allowed_tools: - Read - Glob @@ -41,30 +64,10 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Analysis complete | `[PLANNER:DONE]` | - | Requirements unclear | `[PLANNER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: plan (Task Analysis) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/00-plan.md - - ## User Request - {task} - ## Previous Response (when returned from implement) {previous_response} - ## Instructions Analyze the task and create an implementation plan. **Note:** If returned from implement step (Previous Response exists), @@ -74,47 +77,22 @@ steps: 1. Understand the requirements 2. Identify impact scope 3. Decide implementation approach - - **Report output:** Output to the `Report File` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - - **Report format:** - ```markdown - # Task Plan - - ## Original Request - {User's request as-is} - - ## Analysis Results - - ### Objective - {What needs to be achieved} - - ### Scope - {Impact scope} - - ### Implementation Approach - {How to proceed} - - ## Clarifications Needed (if any) - - {Unclear points or items requiring confirmation} - ``` - - Output [PLANNER:DONE] when complete. - Output [PLANNER:BLOCKED] if requirements are unclear. pass_previous_response: true - transitions: - - condition: done - next_step: implement - - condition: blocked - next_step: ABORT + rules: + - condition: Task analysis and planning is complete + next: implement + - condition: Requirements are unclear and planning cannot proceed + next: ABORT # =========================================== # Phase 1: Implementation # =========================================== - name: implement + edit: true agent: ~/.takt/agents/default/coder.md + report: + - Scope: 01-coder-scope.md + - Decisions: 02-coder-decisions.md allowed_tools: - Read - Glob @@ -124,39 +102,10 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Implementation complete | `[CODER:DONE]` | - | Cannot proceed | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: implement - - Report Directory: .takt/reports/{report_dir}/ - - Report Files: - - Scope: .takt/reports/{report_dir}/01-coder-scope.md - - Decisions: .takt/reports/{report_dir}/02-coder-decisions.md - - ## User Request - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions Follow the plan from the plan step and implement. Refer to the plan report (00-plan.md) and proceed with implementation. - **Report output:** Output to the `Report Files` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - **Scope report format (create at implementation start):** ```markdown # Change Scope Declaration @@ -186,20 +135,46 @@ steps: - **Options Considered**: {List of options} - **Reason**: {Why this option was chosen} ``` - - Include [CODER:DONE] when complete. - Include [CODER:BLOCKED] if you cannot proceed (returns to plan). - transitions: - - condition: done - next_step: cqrs_es_review - - condition: blocked - next_step: plan + rules: + - condition: Implementation is complete + next: ai_review + - condition: Cannot proceed with implementation + next: plan # =========================================== - # Phase 2: CQRS+ES Review + # Phase 2: AI Review # =========================================== - - name: cqrs_es_review - agent: ~/.takt/agents/expert-cqrs/cqrs-es-reviewer.md + - name: ai_review + edit: false + agent: ~/.takt/agents/default/ai-antipattern-reviewer.md + report: + name: 03-ai-review.md + format: | + ```markdown + # AI-Generated Code Review + + ## Result: APPROVE / REJECT + + ## Summary + {One sentence summarizing result} + + ## Verified Items + | Aspect | Result | Notes | + |--------|--------|-------| + | Assumption validity | ✅ | - | + | API/Library existence | ✅ | - | + | Context fit | ✅ | - | + | Scope | ✅ | - | + + ## Issues (if REJECT) + | # | Category | Location | Issue | + |---|----------|----------|-------| + | 1 | Hallucinated API | `src/file.ts:23` | Non-existent method | + ``` + + **Cognitive load reduction rules:** + - No issues → Summary 1 line + check table only (10 lines or less) + - Issues found → + Issues in table format (25 lines or less) allowed_tools: - Read - Glob @@ -207,32 +182,86 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Design is sound | `[CQRS-ES:APPROVE]` | - | Design issues found | `[CQRS-ES:REJECT]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: cqrs_es_review (CQRS+ES Expert Review) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/03-cqrs-es-review.md + Review the code for AI-specific issues: + - Assumption validation + - Plausible but wrong patterns + - Context fit with existing codebase + - Scope creep detection + rules: + - condition: No AI-specific issues found + next: cqrs_es_review + - condition: AI-specific issues detected + next: ai_fix - ## Original User Request - {task} + - name: ai_fix + edit: true + agent: ~/.takt/agents/default/coder.md + allowed_tools: + - Read + - Glob + - Grep + - Edit + - Write + - Bash + - WebSearch + - WebFetch + instruction_template: | + ## AI Review Feedback (This is the latest instruction - prioritize this) + {previous_response} - ## Git Diff - ```diff - {git_diff} - ``` + **Important**: Address the AI Reviewer's feedback. + Focus on: + - Correcting incorrect assumptions + - Fixing plausible-but-wrong implementations + - Aligning with existing codebase patterns + - Removing scope creep + pass_previous_response: true + rules: + - condition: AI Reviewer's issues have been fixed + next: ai_review + - condition: Unable to proceed with fixes + next: plan - ## Instructions + # =========================================== + # Phase 3: CQRS+ES Review + # =========================================== + - name: cqrs_es_review + edit: false + agent: ~/.takt/agents/expert-cqrs/cqrs-es-reviewer.md + report: + name: 04-cqrs-es-review.md + format: | + ```markdown + # CQRS+ES Review + + ## Result: APPROVE / REJECT + + ## Summary + {1-2 sentences summarizing result} + + ## Reviewed Perspectives + | Perspective | Result | Notes | + |-------------|--------|-------| + | Aggregate Design | ✅ | - | + | Event Design | ✅ | - | + | Command/Query Separation | ✅ | - | + | Projections | ✅ | - | + | Eventual Consistency | ✅ | - | + + ## Issues (if REJECT) + | # | Location | Issue | Fix | + |---|----------|-------|-----| + | 1 | `src/file.ts:42` | Issue description | Fix method | + ``` + allowed_tools: + - Read + - Glob + - Grep + - Write + - WebSearch + - WebFetch + instruction_template: | Review the changes above from the CQRS (Command Query Responsibility Segregation) and Event Sourcing perspective. @@ -245,45 +274,14 @@ steps: **Note**: If this project does not use CQRS+ES patterns, review from a general domain design perspective. - - **Report output:** Output to the `Report File` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - - **Report format:** - ```markdown - # CQRS+ES Review - - ## Result: APPROVE / REJECT - - ## Summary - {1-2 sentences summarizing result} - - ## Reviewed Perspectives - | Perspective | Result | Notes | - |-------------|--------|-------| - | Aggregate Design | ✅ | - | - | Event Design | ✅ | - | - | Command/Query Separation | ✅ | - | - | Projections | ✅ | - | - | Eventual Consistency | ✅ | - | - - ## Issues (if REJECT) - | # | Location | Issue | Fix | - |---|----------|-------|-----| - | 1 | `src/file.ts:42` | Issue description | Fix method | - ``` - - Include: - - [CQRS-ES:APPROVE] if CQRS+ES design is sound - - [CQRS-ES:REJECT] if design issues found (list specific issues) - transitions: - - condition: approved - next_step: frontend_review - - condition: rejected - next_step: fix_cqrs_es + rules: + - condition: CQRS+ES design is sound with no issues + next: frontend_review + - condition: CQRS+ES design issues found + next: fix_cqrs_es - name: fix_cqrs_es + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -294,31 +292,10 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Fix complete | `[CODER:DONE]` | - | Cannot proceed | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: fix_cqrs_es - ## CQRS+ES Review Feedback (This is the latest instruction - prioritize this) {previous_response} - ## Original User Request (Initial request - for reference) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **Important**: Fix the issues pointed out by the CQRS+ES expert. Areas of concern: @@ -327,21 +304,44 @@ steps: - Command/Query separation - Projections - Eventual consistency - - Include [CODER:DONE] when complete. - Include [CODER:BLOCKED] if unable to proceed. pass_previous_response: true - transitions: - - condition: done - next_step: cqrs_es_review - - condition: blocked - next_step: plan + rules: + - condition: CQRS+ES issues have been fixed + next: cqrs_es_review + - condition: Unable to proceed with fixes + next: plan # =========================================== - # Phase 3: Frontend Review + # Phase 4: Frontend Review # =========================================== - name: frontend_review + edit: false agent: ~/.takt/agents/expert/frontend-reviewer.md + report: + name: 05-frontend-review.md + format: | + ```markdown + # Frontend Review + + ## Result: APPROVE / REJECT + + ## Summary + {1-2 sentences summarizing result} + + ## Reviewed Perspectives + | Perspective | Result | Notes | + |-------------|--------|-------| + | Component Design | ✅ | - | + | State Management | ✅ | - | + | Performance | ✅ | - | + | Accessibility | ✅ | - | + | Type Safety | ✅ | - | + + ## Issues (if REJECT) + | # | Location | Issue | Fix | + |---|----------|-------|-----| + | 1 | `src/file.tsx:42` | Issue description | Fix method | + ``` allowed_tools: - Read - Glob @@ -349,32 +349,7 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Frontend design is sound | `[FRONTEND:APPROVE]` | - | Design issues found | `[FRONTEND:REJECT]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: frontend_review (Frontend Expert Review) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/04-frontend-review.md - - ## Original User Request - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions Review the changes above from the frontend development perspective. **Review Criteria:** @@ -386,46 +361,15 @@ steps: - TypeScript type safety **Note**: If this project does not include frontend code, - output [FRONTEND:APPROVE] and proceed. - - **Report output:** Output to the `Report File` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - - **Report format:** - ```markdown - # Frontend Review - - ## Result: APPROVE / REJECT - - ## Summary - {1-2 sentences summarizing result} - - ## Reviewed Perspectives - | Perspective | Result | Notes | - |-------------|--------|-------| - | Component Design | ✅ | - | - | State Management | ✅ | - | - | Performance | ✅ | - | - | Accessibility | ✅ | - | - | Type Safety | ✅ | - | - - ## Issues (if REJECT) - | # | Location | Issue | Fix | - |---|----------|-------|-----| - | 1 | `src/file.tsx:42` | Issue description | Fix method | - ``` - - Include: - - [FRONTEND:APPROVE] if frontend design is sound - - [FRONTEND:REJECT] if design issues found (list specific issues) - transitions: - - condition: approved - next_step: ai_review - - condition: rejected - next_step: fix_frontend + approve and proceed to the next step. + rules: + - condition: Frontend design is sound with no issues + next: security_review + - condition: Frontend design issues found + next: fix_frontend - name: fix_frontend + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -436,31 +380,10 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Fix complete | `[CODER:DONE]` | - | Cannot proceed | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: fix_frontend - ## Frontend Review Feedback (This is the latest instruction - prioritize this) {previous_response} - ## Original User Request (Initial request - for reference) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **Important**: Fix the issues pointed out by the frontend expert. Areas of concern: @@ -469,157 +392,45 @@ steps: - Performance - Accessibility - Type safety - - Include [CODER:DONE] when complete. - Include [CODER:BLOCKED] if unable to proceed. pass_previous_response: true - transitions: - - condition: done - next_step: frontend_review - - condition: blocked - next_step: plan - - # =========================================== - # Phase 4: AI Review - # =========================================== - - name: ai_review - agent: ~/.takt/agents/default/ai-antipattern-reviewer.md - allowed_tools: - - Read - - Glob - - Grep - - Write - - WebSearch - - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | No AI-specific issues | `[AI_REVIEW:APPROVE]` | - | Issues found | `[AI_REVIEW:REJECT]` | - instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: ai_review (AI-Generated Code Review) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/05-ai-review.md - - ## Original User Request (Initial request from workflow start) - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions - Review the code for AI-specific issues: - - Assumption validation - - Plausible but wrong patterns - - Context fit with existing codebase - - Scope creep detection - - **Report output:** Output to the `Report File` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - - **Report format:** - ```markdown - # AI-Generated Code Review - - ## Result: APPROVE / REJECT - - ## Summary - {One sentence summarizing result} - - ## Verified Items - | Aspect | Result | Notes | - |--------|--------|-------| - | Assumption validity | ✅ | - | - | API/Library existence | ✅ | - | - | Context fit | ✅ | - | - | Scope | ✅ | - | - - ## Issues (if REJECT) - | # | Category | Location | Issue | - |---|----------|----------|-------| - | 1 | Hallucinated API | `src/file.ts:23` | Non-existent method | - ``` - - **Cognitive load reduction rules:** - - No issues → Summary 1 line + check table only (10 lines or less) - - Issues found → + Issues in table format (25 lines or less) - - Include: - - [AI_REVIEW:APPROVE] if no AI-specific issues found - - [AI_REVIEW:REJECT] if issues detected (list specific problems) - transitions: - - condition: approved - next_step: security_review - - condition: rejected - next_step: ai_fix - - - name: ai_fix - agent: ~/.takt/agents/default/coder.md - allowed_tools: - - Read - - Glob - - Grep - - Edit - - Write - - Bash - - WebSearch - - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Fix complete | `[CODER:DONE]` | - | Cannot proceed | `[CODER:BLOCKED]` | - instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: ai_fix - - ## AI Review Feedback (This is the latest instruction - prioritize this) - {previous_response} - - ## Original User Request (Initial request from workflow start - for reference) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions - **Important**: Address the AI Reviewer's feedback. - Focus on: - - Correcting incorrect assumptions - - Fixing plausible-but-wrong implementations - - Aligning with existing codebase patterns - - Removing scope creep - - Include [CODER:DONE] when complete. - Include [CODER:BLOCKED] if unable to proceed. - pass_previous_response: true - transitions: - - condition: done - next_step: ai_review - - condition: blocked - next_step: plan + rules: + - condition: Frontend issues have been fixed + next: frontend_review + - condition: Unable to proceed with fixes + next: plan # =========================================== # Phase 5: Security Review # =========================================== - name: security_review + edit: false agent: ~/.takt/agents/expert/security-reviewer.md + report: + name: 06-security-review.md + format: | + ```markdown + # Security Review + + ## Result: APPROVE / REJECT + + ## Severity: None / Low / Medium / High / Critical + + ## Check Results + | Category | Result | Notes | + |----------|--------|-------| + | Injection | ✅ | - | + | Auth/Authz | ✅ | - | + | Data Protection | ✅ | - | + | Dependencies | ✅ | - | + + ## Vulnerabilities (if REJECT) + | # | Severity | Type | Location | Fix | + |---|----------|------|----------|-----| + | 1 | High | SQLi | `src/db.ts:42` | Use parameterized query | + + ## Warnings (non-blocking) + - {Security recommendations} + ``` allowed_tools: - Read - Glob @@ -627,32 +438,7 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | No security issues | `[SECURITY:APPROVE]` | - | Vulnerabilities found | `[SECURITY:REJECT]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: security_review (Security Expert Review) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/06-security-review.md - - ## Original User Request - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions Review the changes above from the security perspective. **Review Criteria:** @@ -661,46 +447,14 @@ steps: - Sensitive information handling - Encryption appropriateness - OWASP Top 10 - - **Report output:** Output to the `Report File` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - - **Report format:** - ```markdown - # Security Review - - ## Result: APPROVE / REJECT - - ## Severity: None / Low / Medium / High / Critical - - ## Check Results - | Category | Result | Notes | - |----------|--------|-------| - | Injection | ✅ | - | - | Auth/Authz | ✅ | - | - | Data Protection | ✅ | - | - | Dependencies | ✅ | - | - - ## Vulnerabilities (if REJECT) - | # | Severity | Type | Location | Fix | - |---|----------|------|----------|-----| - | 1 | High | SQLi | `src/db.ts:42` | Use parameterized query | - - ## Warnings (non-blocking) - - {Security recommendations} - ``` - - Include: - - [SECURITY:APPROVE] if no security issues found - - [SECURITY:REJECT] if vulnerabilities found (list specific issues with severity) - transitions: - - condition: approved - next_step: qa_review - - condition: rejected - next_step: fix_security + rules: + - condition: No security issues found + next: qa_review + - condition: Security vulnerabilities detected + next: fix_security - name: fix_security + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -711,32 +465,10 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Minor fix complete | `[CODER:DONE]` | - | Major fix (restart from CQRS+ES) | `[CODER:REJECT]` | - | Cannot proceed | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: fix_security - ## Security Review Feedback (This is the latest instruction - prioritize this) {previous_response} - ## Original User Request (Initial request - for reference) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **Important**: Fix the issues pointed out by the security expert. Security issues should be addressed with highest priority. @@ -747,28 +479,52 @@ steps: - Encryption issues ## Completion: Determine Change Impact - When fix is complete, judge the **impact scope of changes** and output the appropriate tag: + When fix is complete, judge the **impact scope of changes**: - - `[CODER:DONE]` - Minor fix (re-run security review only) + - Minor fix (re-run security review only) - Examples: Add validation, add escaping, configuration changes - - `[CODER:REJECT]` - Major fix (restart from CQRS+ES review) + - Major fix (restart from CQRS+ES review) - Examples: Data flow changes, API design changes, auth method changes, domain model changes - - Include [CODER:BLOCKED] if unable to proceed. pass_previous_response: true - transitions: - - condition: done - next_step: security_review - - condition: rejected - next_step: cqrs_es_review - - condition: blocked - next_step: plan + rules: + - condition: Minor security fix is complete + next: security_review + - condition: Major fix applied requiring CQRS+ES re-review + next: cqrs_es_review + - condition: Unable to proceed with fixes + next: plan # =========================================== # Phase 6: QA Review # =========================================== - name: qa_review + edit: false agent: ~/.takt/agents/expert/qa-reviewer.md + report: + name: 07-qa-review.md + format: | + ```markdown + # QA Review + + ## Result: APPROVE / REJECT + + ## Summary + {1-2 sentences summarizing result} + + ## Reviewed Perspectives + | Perspective | Result | Notes | + |-------------|--------|-------| + | Test Coverage | ✅ | - | + | Test Quality | ✅ | - | + | Error Handling | ✅ | - | + | Documentation | ✅ | - | + | Maintainability | ✅ | - | + + ## Issues (if REJECT) + | # | Category | Issue | Fix | + |---|----------|-------|-----| + | 1 | Testing | Issue description | Fix method | + ``` allowed_tools: - Read - Glob @@ -776,32 +532,7 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Quality standards met | `[QA:APPROVE]` | - | Quality issues found | `[QA:REJECT]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: qa_review (QA Expert Review) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/07-qa-review.md - - ## Original User Request - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions Review the changes above from the quality assurance perspective. **Review Criteria:** @@ -811,45 +542,14 @@ steps: - Error handling - Logging and monitoring - Maintainability - - **Report output:** Output to the `Report File` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - - **Report format:** - ```markdown - # QA Review - - ## Result: APPROVE / REJECT - - ## Summary - {1-2 sentences summarizing result} - - ## Reviewed Perspectives - | Perspective | Result | Notes | - |-------------|--------|-------| - | Test Coverage | ✅ | - | - | Test Quality | ✅ | - | - | Error Handling | ✅ | - | - | Documentation | ✅ | - | - | Maintainability | ✅ | - | - - ## Issues (if REJECT) - | # | Category | Issue | Fix | - |---|----------|-------|-----| - | 1 | Testing | Issue description | Fix method | - ``` - - Include: - - [QA:APPROVE] if quality standards are met - - [QA:REJECT] if quality issues found (list specific issues) - transitions: - - condition: approved - next_step: supervise - - condition: rejected - next_step: fix_qa + rules: + - condition: Quality standards are met + next: supervise + - condition: Quality issues found + next: fix_qa - name: fix_qa + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -860,33 +560,10 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Minor fix complete | `[CODER:DONE]` | - | Security-impacting fix | `[CODER:IMPROVE]` | - | Major fix (restart from CQRS+ES) | `[CODER:REJECT]` | - | Cannot proceed | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: fix_qa - ## QA Review Feedback (This is the latest instruction - prioritize this) {previous_response} - ## Original User Request (Initial request - for reference) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **Important**: Fix the issues pointed out by the QA expert. Areas of concern: @@ -897,32 +574,34 @@ steps: - Code quality ## Completion: Determine Change Impact - When fix is complete, judge the **impact scope of changes** and output the appropriate tag: + When fix is complete, judge the **impact scope of changes**: - - `[CODER:DONE]` - Minor fix (re-run QA review only) + - Minor fix (re-run QA review only) - Examples: Add tests, add documentation, add logs, add comments - - `[CODER:IMPROVE]` - Security-impacting fix (restart from security review) + - Security-impacting fix (restart from security review) - Examples: Error handling changes (error message content changes), input validation changes - - `[CODER:REJECT]` - Major fix (restart from CQRS+ES review) + - Major fix (restart from CQRS+ES review) - Examples: Business logic changes, data model changes, API changes - - Include [CODER:BLOCKED] if unable to proceed. pass_previous_response: true - transitions: - - condition: done - next_step: qa_review - - condition: improve - next_step: security_review - - condition: rejected - next_step: cqrs_es_review - - condition: blocked - next_step: plan + rules: + - condition: Minor QA fix is complete + next: qa_review + - condition: Security-impacting fix applied + next: security_review + - condition: Major fix applied requiring CQRS+ES re-review + next: cqrs_es_review + - condition: Unable to proceed with fixes + next: plan # =========================================== # Phase 7: Supervision # =========================================== - name: supervise + edit: false agent: ~/.takt/agents/expert/supervisor.md + report: + - Validation: 08-supervisor-validation.md + - Summary: summary.md allowed_tools: - Read - Glob @@ -930,42 +609,15 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Ready to merge | `[SUPERVISOR:APPROVE]` | - | Issues found | `[SUPERVISOR:REJECT]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: supervise (Final Review) - - Report Directory: .takt/reports/{report_dir}/ - - Report Files: - - Validation: .takt/reports/{report_dir}/08-supervisor-validation.md - - Summary: .takt/reports/{report_dir}/summary.md - - ## Original User Request - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - ## Previous Reviews Summary Reaching this step means all the following reviews have been APPROVED: + - AI Review: APPROVED - CQRS+ES Review: APPROVED - Frontend Review: APPROVED - - AI Review: APPROVED - Security Review: APPROVED - QA Review: APPROVED - ## Instructions Run tests, verify the build, and perform final approval. **Workflow Overall Review:** @@ -976,10 +628,6 @@ steps: **Review Reports:** Read all reports in Report Directory and check for any unaddressed improvement suggestions. - **Report output:** Output to the `Report Files` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - **Validation report format:** ```markdown # Final Validation Results @@ -1022,9 +670,9 @@ steps: ## Review Results | Review | Result | |--------|--------| + | AI Review | ✅ APPROVE | | CQRS+ES | ✅ APPROVE | | Frontend | ✅ APPROVE | - | AI Review | ✅ APPROVE | | Security | ✅ APPROVE | | QA | ✅ APPROVE | | Supervisor | ✅ APPROVE | @@ -1035,17 +683,14 @@ steps: npm run build ``` ``` - - Output: - - [SUPERVISOR:APPROVE] if ready to merge - - [SUPERVISOR:REJECT] if issues found (specify the issues) - transitions: - - condition: approved - next_step: COMPLETE - - condition: rejected - next_step: fix_supervisor + rules: + - condition: All validations pass and ready to merge + next: COMPLETE + - condition: Issues detected during final review + next: fix_supervisor - name: fix_supervisor + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -1056,41 +701,17 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Fix complete | `[CODER:DONE]` | - | Cannot proceed | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: fix_supervisor - ## Supervisor Feedback (This is the latest instruction - prioritize this) {previous_response} - ## Original User Request (Initial request - for reference) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **Important**: Fix the issues pointed out by the supervisor. The supervisor has identified issues from a big-picture perspective. Address items in priority order. - - Include [CODER:DONE] when complete. - Include [CODER:BLOCKED] if unable to proceed. pass_previous_response: true - transitions: - - condition: done - next_step: supervise - - condition: blocked - next_step: plan + rules: + - condition: Supervisor's issues have been fixed + next: supervise + - condition: Unable to proceed with fixes + next: plan diff --git a/resources/global/en/workflows/expert.yaml b/resources/global/en/workflows/expert.yaml index fb24f92..a7dd490 100644 --- a/resources/global/en/workflows/expert.yaml +++ b/resources/global/en/workflows/expert.yaml @@ -2,28 +2,39 @@ # Review workflow with Architecture, Frontend, Security, and QA experts # # Flow: -# plan -> implement -> architect_review -> frontend_review -> ai_review -> security_review -> qa_review -> supervise -> COMPLETE -# ↓ ↓ ↓ ↓ ↓ ↓ -# fix_architect fix_frontend ai_fix fix_security fix_qa fix_supervisor +# plan -> implement -> ai_review -> architect_review -> frontend_review -> security_review -> qa_review -> supervise -> COMPLETE +# ↓ ↓ ↓ ↓ ↓ ↓ ↓ +# ai_fix fix_architect fix_frontend fix_security fix_qa fix_supervisor +# +# AI review runs immediately after implementation to catch AI-specific issues early, +# before expert reviews begin. # # Fix destination is determined by Coder based on change impact: -# - fix_security: MINOR→security_review, MAJOR→architect_review -# - fix_qa: MINOR→qa_review, SECURITY→security_review, MAJOR→architect_review +# - fix_security: MINOR->security_review, MAJOR->architect_review +# - fix_qa: MINOR->qa_review, SECURITY->security_review, MAJOR->architect_review # -# Template Variables: -# {iteration} - Workflow-wide turn count (total steps executed across all agents) -# {max_iterations} - Maximum iterations allowed for the workflow -# {step_iteration} - Per-step iteration count (how many times THIS step has been executed) -# {task} - Original user request -# {previous_response} - Output from the previous step -# {git_diff} - Current uncommitted changes (git diff) -# {user_inputs} - Accumulated user inputs during workflow -# {report_dir} - Report directory name (e.g., "20250126-143052-task-summary") +# Boilerplate sections (Workflow Context, User Request, Previous Response, +# Additional User Inputs, Instructions heading) are auto-injected by buildInstruction(). +# Only step-specific content belongs in instruction_template. +# +# Template Variables (available in instruction_template): +# {iteration} - Workflow-wide turn count (total steps executed across all agents) +# {max_iterations} - Maximum iterations allowed for the workflow +# {step_iteration} - Per-step iteration count (how many times THIS step has been executed) +# {previous_response} - Output from the previous step (only when pass_previous_response: true) +# {report_dir} - Report directory name (e.g., "20250126-143052-task-summary") +# +# Step-level Fields: +# report: - Report file(s) for the step (auto-injected as Report File/Files in Workflow Context) +# Single: report: 00-plan.md +# Multiple: report: +# - Scope: 01-coder-scope.md +# - Decisions: 02-coder-decisions.md name: expert description: Architecture, Frontend, Security, QA Expert Review -max_iterations: 20 +max_iterations: 30 initial_step: plan @@ -32,7 +43,31 @@ steps: # Phase 0: Planning # =========================================== - name: plan + edit: false agent: ~/.takt/agents/default/planner.md + report: + name: 00-plan.md + format: | + ```markdown + # Task Plan + + ## Original Request + {User's request as-is} + + ## Analysis Results + + ### Objective + {What needs to be achieved} + + ### Scope + {Impact scope} + + ### Implementation Approach + {How to proceed} + + ## Clarifications Needed (if any) + - {Unclear points or items requiring confirmation} + ``` allowed_tools: - Read - Glob @@ -41,30 +76,10 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Analysis complete | `[PLANNER:DONE]` | - | Requirements unclear | `[PLANNER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: plan (Task Analysis) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/00-plan.md - - ## User Request - {task} - ## Previous Response (when returned from implement) {previous_response} - ## Instructions Analyze the task and create an implementation plan. **Note:** If returned from implement step (Previous Response exists), @@ -74,47 +89,22 @@ steps: 1. Understand the requirements 2. Identify impact scope 3. Decide implementation approach - - **Report output:** Output to the `Report File` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - - **Report format:** - ```markdown - # Task Plan - - ## Original Request - {User's request as-is} - - ## Analysis Results - - ### Objective - {What needs to be achieved} - - ### Scope - {Impact scope} - - ### Implementation Approach - {How to proceed} - - ## Clarifications Needed (if any) - - {Unclear points or items requiring confirmation} - ``` - - Output [PLANNER:DONE] when complete. - Output [PLANNER:BLOCKED] if requirements are unclear. pass_previous_response: true - transitions: - - condition: done - next_step: implement - - condition: blocked - next_step: ABORT + rules: + - condition: Task analysis and planning is complete + next: implement + - condition: Requirements are unclear and planning cannot proceed + next: ABORT # =========================================== # Phase 1: Implementation # =========================================== - name: implement + edit: true agent: ~/.takt/agents/default/coder.md + report: + - Scope: 01-coder-scope.md + - Decisions: 02-coder-decisions.md allowed_tools: - Read - Glob @@ -124,39 +114,10 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Implementation complete | `[CODER:DONE]` | - | Cannot proceed | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: implement - - Report Directory: .takt/reports/{report_dir}/ - - Report Files: - - Scope: .takt/reports/{report_dir}/01-coder-scope.md - - Decisions: .takt/reports/{report_dir}/02-coder-decisions.md - - ## User Request - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions Follow the plan from the plan step and implement. Refer to the plan report (00-plan.md) and proceed with implementation. - **Report output:** Output to the `Report Files` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - **Scope report format (create at implementation start):** ```markdown # Change Scope Declaration @@ -186,20 +147,46 @@ steps: - **Options Considered**: {List of options} - **Reason**: {Why this option was chosen} ``` - - Include [CODER:DONE] when complete. - Include [CODER:BLOCKED] if you cannot proceed (returns to plan). - transitions: - - condition: done - next_step: architect_review - - condition: blocked - next_step: plan + rules: + - condition: Implementation is complete + next: ai_review + - condition: Cannot proceed with implementation + next: plan # =========================================== - # Phase 2: Architecture Review + # Phase 2: AI Review # =========================================== - - name: architect_review - agent: ~/.takt/agents/default/architect.md + - name: ai_review + edit: false + agent: ~/.takt/agents/default/ai-antipattern-reviewer.md + report: + name: 03-ai-review.md + format: | + ```markdown + # AI-Generated Code Review + + ## Result: APPROVE / REJECT + + ## Summary + {One sentence summarizing result} + + ## Verified Items + | Aspect | Result | Notes | + |--------|--------|-------| + | Assumption validity | ✅ | - | + | API/Library existence | ✅ | - | + | Context fit | ✅ | - | + | Scope | ✅ | - | + + ## Issues (if REJECT) + | # | Category | Location | Issue | + |---|----------|----------|-------| + | 1 | Hallucinated API | `src/file.ts:23` | Non-existent method | + ``` + + **Cognitive load reduction rules:** + - No issues -> Summary 1 line + check table only (10 lines or less) + - Issues found -> + Issues in table format (25 lines or less) allowed_tools: - Read - Glob @@ -207,44 +194,93 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - - - ## Judgment Criteria - - | Situation | Judgment | - |-----------|----------| - | Structural issues | REJECT | - | Design principle violations | REJECT | - | Missing call chain wiring | REJECT | - | Insufficient tests | REJECT | - | Minor improvements needed | IMPROVE | - | No issues | APPROVE | - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | No issues | `[ARCHITECT:APPROVE]` | - | Minor improvements needed | `[ARCHITECT:IMPROVE]` | - | Structural fixes required | `[ARCHITECT:REJECT]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: architect_review (Architecture Review) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/03-architect-review.md + Review the code for AI-specific issues: + - Assumption validation + - Plausible but wrong patterns + - Context fit with existing codebase + - Scope creep detection + rules: + - condition: No AI-specific issues found + next: architect_review + - condition: AI-specific issues detected + next: ai_fix - ## Original User Request - {task} + - name: ai_fix + edit: true + agent: ~/.takt/agents/default/coder.md + allowed_tools: + - Read + - Glob + - Grep + - Edit + - Write + - Bash + - WebSearch + - WebFetch + instruction_template: | + ## AI Review Feedback (This is the latest instruction - prioritize this) + {previous_response} - ## Git Diff - ```diff - {git_diff} - ``` + **Important**: Address the AI Reviewer's feedback. + Focus on: + - Correcting incorrect assumptions + - Fixing plausible-but-wrong implementations + - Aligning with existing codebase patterns + - Removing scope creep + pass_previous_response: true + rules: + - condition: AI Reviewer's issues have been fixed + next: ai_review + - condition: Unable to proceed with fixes + next: plan - ## Instructions + # =========================================== + # Phase 3: Architecture Review + # =========================================== + - name: architect_review + edit: false + agent: ~/.takt/agents/default/architecture-reviewer.md + report: + name: 04-architect-review.md + format: | + ```markdown + # Architecture Review + + ## Result: APPROVE / IMPROVE / REJECT + + ## Summary + {1-2 sentences summarizing result} + + ## Reviewed Aspects + - [x] Structure/Design + - [x] Code Quality + - [x] Change Scope + - [x] Test Coverage + - [x] Dead Code + - [x] Call Chain Verification + + ## Issues (if REJECT) + | # | Location | Issue | Fix | + |---|----------|-------|-----| + | 1 | `src/file.ts:42` | Issue description | Fix method | + + ## Improvement Suggestions (optional - non-blocking) + - {Future improvement suggestions} + ``` + + **Cognitive load reduction rules:** + - APPROVE + no issues -> Summary only (5 lines or less) + - APPROVE + minor suggestions -> Summary + suggestions (15 lines or less) + - REJECT -> Issues in table format (30 lines or less) + allowed_tools: + - Read + - Glob + - Grep + - Write + - WebSearch + - WebFetch + instruction_template: | Focus on **architecture and design** review. **Review Criteria:** @@ -254,50 +290,16 @@ steps: - Test coverage - Dead code - Call chain verification - - **Report output:** Output to the `Report File` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - - **Report format:** - ```markdown - # Architecture Review - - ## Result: APPROVE / IMPROVE / REJECT - - ## Summary - {1-2 sentences summarizing result} - - ## Reviewed Aspects - - [x] Structure/Design - - [x] Code Quality - - [x] Change Scope - - [x] Test Coverage - - [x] Dead Code - - [x] Call Chain Verification - - ## Issues (if REJECT) - | # | Location | Issue | Fix | - |---|----------|-------|-----| - | 1 | `src/file.ts:42` | Issue description | Fix method | - - ## Improvement Suggestions (optional - non-blocking) - - {Future improvement suggestions} - ``` - - **Cognitive load reduction rules:** - - APPROVE + no issues → Summary only (5 lines or less) - - APPROVE + minor suggestions → Summary + suggestions (15 lines or less) - - REJECT → Issues in table format (30 lines or less) - transitions: - - condition: approved - next_step: frontend_review - - condition: improve - next_step: fix_architect - - condition: rejected - next_step: fix_architect + rules: + - condition: No architecture or design issues found + next: frontend_review + - condition: Minor improvements needed but no structural issues + next: fix_architect + - condition: Structural issues found that require fixes + next: fix_architect - name: fix_architect + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -309,49 +311,51 @@ steps: - WebSearch - WebFetch permission_mode: acceptEdits - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Fix complete | `[CODER:DONE]` | - | Cannot proceed | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: fix_architect - ## Architect Feedback (This is the latest instruction - prioritize this) {previous_response} - ## Original User Request (Initial request - for reference) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **Important**: Address the Architect's feedback. "Original User Request" is for reference; it's not the latest instruction. Review the session conversation history and fix the Architect's issues. - - Include [CODER:DONE] when complete. - Include [CODER:BLOCKED] if unable to proceed. pass_previous_response: true - transitions: - - condition: done - next_step: architect_review - - condition: blocked - next_step: plan + rules: + - condition: Architect's issues have been fixed + next: architect_review + - condition: Unable to proceed with fixes + next: plan # =========================================== - # Phase 3: Frontend Review + # Phase 4: Frontend Review # =========================================== - name: frontend_review + edit: false agent: ~/.takt/agents/expert/frontend-reviewer.md + report: + name: 05-frontend-review.md + format: | + ```markdown + # Frontend Review + + ## Result: APPROVE / REJECT + + ## Summary + {1-2 sentences summarizing result} + + ## Reviewed Perspectives + | Perspective | Result | Notes | + |-------------|--------|-------| + | Component Design | ✅ | - | + | State Management | ✅ | - | + | Performance | ✅ | - | + | Accessibility | ✅ | - | + | Type Safety | ✅ | - | + + ## Issues (if REJECT) + | # | Location | Issue | Fix | + |---|----------|-------|-----| + | 1 | `src/file.tsx:42` | Issue description | Fix method | + ``` allowed_tools: - Read - Glob @@ -359,33 +363,8 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Frontend design is sound | `[FRONTEND:APPROVE]` | - | Design issues found | `[FRONTEND:REJECT]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: frontend_review (Frontend Expert Review) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/04-frontend-review.md - - ## Original User Request - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions - Review the changes above from the frontend development perspective. + Review the changes from the frontend development perspective. **Review Criteria:** - Component design (separation of concerns, granularity) @@ -396,46 +375,15 @@ steps: - TypeScript type safety **Note**: If this project does not include frontend code, - output [FRONTEND:APPROVE] and proceed. - - **Report output:** Output to the `Report File` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - - **Report format:** - ```markdown - # Frontend Review - - ## Result: APPROVE / REJECT - - ## Summary - {1-2 sentences summarizing result} - - ## Reviewed Perspectives - | Perspective | Result | Notes | - |-------------|--------|-------| - | Component Design | ✅ | - | - | State Management | ✅ | - | - | Performance | ✅ | - | - | Accessibility | ✅ | - | - | Type Safety | ✅ | - | - - ## Issues (if REJECT) - | # | Location | Issue | Fix | - |---|----------|-------|-----| - | 1 | `src/file.tsx:42` | Issue description | Fix method | - ``` - - Include: - - [FRONTEND:APPROVE] if frontend design is sound - - [FRONTEND:REJECT] if design issues found (list specific issues) - transitions: - - condition: approved - next_step: ai_review - - condition: rejected - next_step: fix_frontend + approve and proceed to the next step. + rules: + - condition: Frontend design is sound with no issues + next: security_review + - condition: Frontend design issues found + next: fix_frontend - name: fix_frontend + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -446,31 +394,10 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Fix complete | `[CODER:DONE]` | - | Cannot proceed | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: fix_frontend - ## Frontend Review Feedback (This is the latest instruction - prioritize this) {previous_response} - ## Original User Request (Initial request - for reference) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **Important**: Fix the issues pointed out by the frontend expert. Areas of concern: @@ -479,157 +406,45 @@ steps: - Performance - Accessibility - Type safety - - Include [CODER:DONE] when complete. - Include [CODER:BLOCKED] if unable to proceed. pass_previous_response: true - transitions: - - condition: done - next_step: frontend_review - - condition: blocked - next_step: plan - - # =========================================== - # Phase 4: AI Review - # =========================================== - - name: ai_review - agent: ~/.takt/agents/default/ai-antipattern-reviewer.md - allowed_tools: - - Read - - Glob - - Grep - - Write - - WebSearch - - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | No AI-specific issues | `[AI_REVIEW:APPROVE]` | - | Issues found | `[AI_REVIEW:REJECT]` | - instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: ai_review (AI-Generated Code Review) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/05-ai-review.md - - ## Original User Request (Initial request from workflow start) - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions - Review the code for AI-specific issues: - - Assumption validation - - Plausible but wrong patterns - - Context fit with existing codebase - - Scope creep detection - - **Report output:** Output to the `Report File` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - - **Report format:** - ```markdown - # AI-Generated Code Review - - ## Result: APPROVE / REJECT - - ## Summary - {One sentence summarizing result} - - ## Verified Items - | Aspect | Result | Notes | - |--------|--------|-------| - | Assumption validity | ✅ | - | - | API/Library existence | ✅ | - | - | Context fit | ✅ | - | - | Scope | ✅ | - | - - ## Issues (if REJECT) - | # | Category | Location | Issue | - |---|----------|----------|-------| - | 1 | Hallucinated API | `src/file.ts:23` | Non-existent method | - ``` - - **Cognitive load reduction rules:** - - No issues → Summary 1 line + check table only (10 lines or less) - - Issues found → + Issues in table format (25 lines or less) - - Include: - - [AI_REVIEW:APPROVE] if no AI-specific issues found - - [AI_REVIEW:REJECT] if issues detected (list specific problems) - transitions: - - condition: approved - next_step: security_review - - condition: rejected - next_step: ai_fix - - - name: ai_fix - agent: ~/.takt/agents/default/coder.md - allowed_tools: - - Read - - Glob - - Grep - - Edit - - Write - - Bash - - WebSearch - - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Fix complete | `[CODER:DONE]` | - | Cannot proceed | `[CODER:BLOCKED]` | - instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: ai_fix - - ## AI Review Feedback (This is the latest instruction - prioritize this) - {previous_response} - - ## Original User Request (Initial request from workflow start - for reference) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions - **Important**: Address the AI Reviewer's feedback. - Focus on: - - Correcting incorrect assumptions - - Fixing plausible-but-wrong implementations - - Aligning with existing codebase patterns - - Removing scope creep - - Include [CODER:DONE] when complete. - Include [CODER:BLOCKED] if unable to proceed. - pass_previous_response: true - transitions: - - condition: done - next_step: ai_review - - condition: blocked - next_step: plan + rules: + - condition: Frontend issues have been fixed + next: frontend_review + - condition: Unable to proceed with fixes + next: plan # =========================================== # Phase 5: Security Review # =========================================== - name: security_review + edit: false agent: ~/.takt/agents/expert/security-reviewer.md + report: + name: 06-security-review.md + format: | + ```markdown + # Security Review + + ## Result: APPROVE / REJECT + + ## Severity: None / Low / Medium / High / Critical + + ## Check Results + | Category | Result | Notes | + |----------|--------|-------| + | Injection | ✅ | - | + | Auth/Authz | ✅ | - | + | Data Protection | ✅ | - | + | Dependencies | ✅ | - | + + ## Vulnerabilities (if REJECT) + | # | Severity | Type | Location | Fix | + |---|----------|------|----------|-----| + | 1 | High | SQLi | `src/db.ts:42` | Use parameterized query | + + ## Warnings (non-blocking) + - {Security recommendations} + ``` allowed_tools: - Read - Glob @@ -637,33 +452,8 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | No security issues | `[SECURITY:APPROVE]` | - | Vulnerabilities found | `[SECURITY:REJECT]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: security_review (Security Expert Review) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/06-security-review.md - - ## Original User Request - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions - Review the changes above from the security perspective. + Review the changes from the security perspective. **Review Criteria:** - Injection attacks (SQL, command, XSS) @@ -671,46 +461,14 @@ steps: - Sensitive information handling - Encryption appropriateness - OWASP Top 10 - - **Report output:** Output to the `Report File` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - - **Report format:** - ```markdown - # Security Review - - ## Result: APPROVE / REJECT - - ## Severity: None / Low / Medium / High / Critical - - ## Check Results - | Category | Result | Notes | - |----------|--------|-------| - | Injection | ✅ | - | - | Auth/Authz | ✅ | - | - | Data Protection | ✅ | - | - | Dependencies | ✅ | - | - - ## Vulnerabilities (if REJECT) - | # | Severity | Type | Location | Fix | - |---|----------|------|----------|-----| - | 1 | High | SQLi | `src/db.ts:42` | Use parameterized query | - - ## Warnings (non-blocking) - - {Security recommendations} - ``` - - Include: - - [SECURITY:APPROVE] if no security issues found - - [SECURITY:REJECT] if vulnerabilities found (list specific issues with severity) - transitions: - - condition: approved - next_step: qa_review - - condition: rejected - next_step: fix_security + rules: + - condition: No security issues found + next: qa_review + - condition: Security vulnerabilities detected + next: fix_security - name: fix_security + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -721,32 +479,10 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Minor fix complete | `[CODER:DONE]` | - | Major fix (restart from Architecture) | `[CODER:REJECT]` | - | Cannot proceed | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: fix_security - ## Security Review Feedback (This is the latest instruction - prioritize this) {previous_response} - ## Original User Request (Initial request - for reference) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **Important**: Fix the issues pointed out by the security expert. Security issues should be addressed with highest priority. @@ -757,28 +493,52 @@ steps: - Encryption issues ## Completion: Determine Change Impact - When fix is complete, judge the **impact scope of changes** and output the appropriate tag: + When fix is complete, judge the **impact scope of changes**: - - `[CODER:DONE]` - Minor fix (re-run security review only) + - Minor fix (re-run security review only) - Examples: Add validation, add escaping, configuration changes - - `[CODER:REJECT]` - Major fix (restart from Architecture review) + - Major fix (restart from Architecture review) - Examples: Data flow changes, API design changes, auth method changes, domain model changes - - Include [CODER:BLOCKED] if unable to proceed. pass_previous_response: true - transitions: - - condition: done - next_step: security_review - - condition: rejected - next_step: architect_review - - condition: blocked - next_step: plan + rules: + - condition: Minor security fix is complete + next: security_review + - condition: Major fix applied requiring architecture re-review + next: architect_review + - condition: Unable to proceed with fixes + next: plan # =========================================== # Phase 6: QA Review # =========================================== - name: qa_review + edit: false agent: ~/.takt/agents/expert/qa-reviewer.md + report: + name: 07-qa-review.md + format: | + ```markdown + # QA Review + + ## Result: APPROVE / REJECT + + ## Summary + {1-2 sentences summarizing result} + + ## Reviewed Perspectives + | Perspective | Result | Notes | + |-------------|--------|-------| + | Test Coverage | ✅ | - | + | Test Quality | ✅ | - | + | Error Handling | ✅ | - | + | Documentation | ✅ | - | + | Maintainability | ✅ | - | + + ## Issues (if REJECT) + | # | Category | Issue | Fix | + |---|----------|-------|-----| + | 1 | Testing | Issue description | Fix method | + ``` allowed_tools: - Read - Glob @@ -786,33 +546,8 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Quality standards met | `[QA:APPROVE]` | - | Quality issues found | `[QA:REJECT]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: qa_review (QA Expert Review) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/07-qa-review.md - - ## Original User Request - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions - Review the changes above from the quality assurance perspective. + Review the changes from the quality assurance perspective. **Review Criteria:** - Test coverage and quality @@ -821,45 +556,14 @@ steps: - Error handling - Logging and monitoring - Maintainability - - **Report output:** Output to the `Report File` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - - **Report format:** - ```markdown - # QA Review - - ## Result: APPROVE / REJECT - - ## Summary - {1-2 sentences summarizing result} - - ## Reviewed Perspectives - | Perspective | Result | Notes | - |-------------|--------|-------| - | Test Coverage | ✅ | - | - | Test Quality | ✅ | - | - | Error Handling | ✅ | - | - | Documentation | ✅ | - | - | Maintainability | ✅ | - | - - ## Issues (if REJECT) - | # | Category | Issue | Fix | - |---|----------|-------|-----| - | 1 | Testing | Issue description | Fix method | - ``` - - Include: - - [QA:APPROVE] if quality standards are met - - [QA:REJECT] if quality issues found (list specific issues) - transitions: - - condition: approved - next_step: supervise - - condition: rejected - next_step: fix_qa + rules: + - condition: Quality standards are met + next: supervise + - condition: Quality issues found + next: fix_qa - name: fix_qa + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -870,33 +574,10 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Minor fix complete | `[CODER:DONE]` | - | Security-impacting fix | `[CODER:IMPROVE]` | - | Major fix (restart from Architecture) | `[CODER:REJECT]` | - | Cannot proceed | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: fix_qa - ## QA Review Feedback (This is the latest instruction - prioritize this) {previous_response} - ## Original User Request (Initial request - for reference) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **Important**: Fix the issues pointed out by the QA expert. Areas of concern: @@ -907,32 +588,34 @@ steps: - Code quality ## Completion: Determine Change Impact - When fix is complete, judge the **impact scope of changes** and output the appropriate tag: + When fix is complete, judge the **impact scope of changes**: - - `[CODER:DONE]` - Minor fix (re-run QA review only) + - Minor fix (re-run QA review only) - Examples: Add tests, add documentation, add logs, add comments - - `[CODER:IMPROVE]` - Security-impacting fix (restart from security review) + - Security-impacting fix (restart from security review) - Examples: Error handling changes (error message content changes), input validation changes - - `[CODER:REJECT]` - Major fix (restart from Architecture review) + - Major fix (restart from Architecture review) - Examples: Business logic changes, data model changes, API changes - - Include [CODER:BLOCKED] if unable to proceed. pass_previous_response: true - transitions: - - condition: done - next_step: qa_review - - condition: improve - next_step: security_review - - condition: rejected - next_step: architect_review - - condition: blocked - next_step: plan + rules: + - condition: Minor QA fix is complete + next: qa_review + - condition: Security-impacting fix applied + next: security_review + - condition: Major fix applied requiring architecture re-review + next: architect_review + - condition: Unable to proceed with fixes + next: plan # =========================================== # Phase 7: Supervision # =========================================== - name: supervise + edit: false agent: ~/.takt/agents/expert/supervisor.md + report: + - Validation: 08-supervisor-validation.md + - Summary: summary.md allowed_tools: - Read - Glob @@ -940,33 +623,7 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Ready to merge | `[SUPERVISOR:APPROVE]` | - | Issues found | `[SUPERVISOR:REJECT]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: supervise (Final Review) - - Report Directory: .takt/reports/{report_dir}/ - - Report Files: - - Validation: .takt/reports/{report_dir}/08-supervisor-validation.md - - Summary: .takt/reports/{report_dir}/summary.md - - ## Original User Request - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - ## Previous Reviews Summary Reaching this step means all the following reviews have been APPROVED: - Architecture Review: APPROVED @@ -975,7 +632,6 @@ steps: - Security Review: APPROVED - QA Review: APPROVED - ## Instructions Run tests, verify the build, and perform final approval. **Workflow Overall Review:** @@ -986,10 +642,6 @@ steps: **Review Reports:** Read all reports in Report Directory and check for any unaddressed improvement suggestions. - **Report output:** Output to the `Report Files` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - **Validation report format:** ```markdown # Final Validation Results @@ -1045,17 +697,14 @@ steps: npm run build ``` ``` - - Output: - - [SUPERVISOR:APPROVE] if ready to merge - - [SUPERVISOR:REJECT] if issues found (specify the issues) - transitions: - - condition: approved - next_step: COMPLETE - - condition: rejected - next_step: fix_supervisor + rules: + - condition: All validations pass and ready to merge + next: COMPLETE + - condition: Issues detected during final review + next: fix_supervisor - name: fix_supervisor + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -1066,41 +715,17 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Fix complete | `[CODER:DONE]` | - | Cannot proceed | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: fix_supervisor - ## Supervisor Feedback (This is the latest instruction - prioritize this) {previous_response} - ## Original User Request (Initial request - for reference) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **Important**: Fix the issues pointed out by the supervisor. The supervisor has identified issues from a big-picture perspective. Address items in priority order. - - Include [CODER:DONE] when complete. - Include [CODER:BLOCKED] if unable to proceed. pass_previous_response: true - transitions: - - condition: done - next_step: supervise - - condition: blocked - next_step: plan + rules: + - condition: Supervisor's issues have been fixed + next: supervise + - condition: Unable to proceed with fixes + next: plan diff --git a/resources/global/en/workflows/magi.yaml b/resources/global/en/workflows/magi.yaml index 5715c3a..03bf470 100644 --- a/resources/global/en/workflows/magi.yaml +++ b/resources/global/en/workflows/magi.yaml @@ -8,7 +8,6 @@ # {step_iteration} - Per-step iteration count (how many times THIS step has been executed) # {task} - Original user request # {previous_response} - Output from the previous step -# {git_diff} - Current uncommitted changes (git diff) # {user_inputs} - Accumulated user inputs during workflow # {report_dir} - Report directory name (e.g., "20250126-143052-task-summary") @@ -26,8 +25,15 @@ steps: - Grep - WebSearch - WebFetch - status_rules_prompt: | + instruction_template: | + # MAGI System Initiated + ## Matter for Deliberation + {task} + + ## Instructions + You are MELCHIOR-1 of the MAGI System. + Analyze the above from the perspective of a scientist/engineer and render your judgment. ## Output Format @@ -44,18 +50,9 @@ steps: Reason: {Reason for approval} ``` - instruction_template: | - # MAGI System Initiated - - ## Matter for Deliberation - {task} - - ## Instructions - You are MELCHIOR-1 of the MAGI System. - Analyze the above from the perspective of a scientist/engineer and render your judgment. - transitions: - - condition: always - next_step: balthasar + rules: + - condition: Judgment completed + next: balthasar - name: balthasar agent: ~/.takt/agents/magi/balthasar.md @@ -65,8 +62,19 @@ steps: - Grep - WebSearch - WebFetch - status_rules_prompt: | + instruction_template: | + # MAGI System Continuing + ## Matter for Deliberation + {task} + + ## MELCHIOR-1's Judgment + {previous_response} + + ## Instructions + You are BALTHASAR-2 of the MAGI System. + Analyze the above from the perspective of a nurturer and render your judgment. + Consider MELCHIOR's judgment as reference, but make your own independent assessment. ## Output Format @@ -83,23 +91,10 @@ steps: Reason: {Reason for approval} ``` - instruction_template: | - # MAGI System Continuing - - ## Matter for Deliberation - {task} - - ## MELCHIOR-1's Judgment - {previous_response} - - ## Instructions - You are BALTHASAR-2 of the MAGI System. - Analyze the above from the perspective of a nurturer and render your judgment. - Consider MELCHIOR's judgment as reference, but make your own independent assessment. pass_previous_response: true - transitions: - - condition: always - next_step: casper + rules: + - condition: Judgment completed + next: casper - name: casper agent: ~/.takt/agents/magi/casper.md @@ -109,8 +104,20 @@ steps: - Grep - WebSearch - WebFetch - status_rules_prompt: | + instruction_template: | + # MAGI System Final Deliberation + ## Matter for Deliberation + {task} + + ## Previous Judgments + {previous_response} + + ## Instructions + You are CASPER-3 of the MAGI System. + Analyze the above from a practical/realistic perspective and render your judgment. + + **Finally, tally the judgments from all three and provide the final conclusion.** ## Output Format @@ -137,21 +144,7 @@ steps: [Reasoning/Summary] ``` - instruction_template: | - # MAGI System Final Deliberation - - ## Matter for Deliberation - {task} - - ## Previous Judgments - {previous_response} - - ## Instructions - You are CASPER-3 of the MAGI System. - Analyze the above from a practical/realistic perspective and render your judgment. - - **Finally, tally the judgments from all three and provide the final conclusion.** pass_previous_response: true - transitions: - - condition: always - next_step: COMPLETE + rules: + - condition: Final judgment completed + next: COMPLETE diff --git a/resources/global/en/workflows/research.yaml b/resources/global/en/workflows/research.yaml index 313c435..762e3b5 100644 --- a/resources/global/en/workflows/research.yaml +++ b/resources/global/en/workflows/research.yaml @@ -12,7 +12,6 @@ # {step_iteration} - Per-step iteration count (how many times THIS step has been executed) # {task} - Original user request # {previous_response} - Output from the previous step -# {git_diff} - Current uncommitted changes (git diff) # {user_inputs} - Accumulated user inputs during workflow # {report_dir} - Report directory name (e.g., "20250126-143052-task-summary") @@ -30,30 +29,6 @@ steps: - Grep - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Plan complete | `[PLANNER:DONE]` | - | Insufficient info | `[PLANNER:BLOCKED]` | - - ### Output Examples - - **DONE case:** - ``` - [PLANNER:DONE] - ``` - - **BLOCKED case:** - ``` - [PLANNER:BLOCKED] - - Clarifications needed: - - {Question 1} - ``` instruction_template: | ## Workflow Status - Iteration: {iteration}/{max_iterations} (workflow-wide) @@ -77,11 +52,11 @@ steps: - If multiple interpretations exist, include all in the research scope - If there is feedback from Supervisor, incorporate it into the plan pass_previous_response: true - transitions: - - condition: done - next_step: dig - - condition: blocked - next_step: ABORT + rules: + - condition: Planning is complete + next: dig + - condition: Insufficient information to create a plan + next: ABORT - name: dig agent: ~/.takt/agents/research/digger.md @@ -91,29 +66,6 @@ steps: - Grep - WebSearch - WebFetch - status_rules_prompt: | - - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Research complete | `[DIGGER:DONE]` | - | Unable to research | `[DIGGER:BLOCKED]` | - - ### Output Examples - - **DONE case:** - ``` - [DIGGER:DONE] - ``` - - **BLOCKED case:** - ``` - [DIGGER:BLOCKED] - - Reason: {Why research was not possible} - ``` instruction_template: | ## Workflow Status - Iteration: {iteration}/{max_iterations} (workflow-wide) @@ -142,11 +94,11 @@ steps: - Codebase search - File reading pass_previous_response: true - transitions: - - condition: done - next_step: supervise - - condition: blocked - next_step: ABORT + rules: + - condition: Research is complete + next: supervise + - condition: Unable to conduct research + next: ABORT - name: supervise agent: ~/.takt/agents/research/supervisor.md @@ -156,39 +108,6 @@ steps: - Grep - WebSearch - WebFetch - status_rules_prompt: | - - - ## Judgment Criteria - - | Situation | Judgment | - |-----------|----------| - | Research results sufficient | APPROVE | - | Research results insufficient | REJECT | - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Research complete, results sufficient | `[SUPERVISOR:APPROVE]` | - | Insufficient, restart from planning | `[SUPERVISOR:REJECT]` | - - ### Output Examples - - **APPROVE case:** - ``` - [SUPERVISOR:APPROVE] - - Research results adequately answer the original request. - ``` - - **REJECT case:** - ``` - [SUPERVISOR:REJECT] - - Missing: - - {Specific missing items} - ``` instruction_template: | ## Workflow Status - Iteration: {iteration}/{max_iterations} (workflow-wide) @@ -206,10 +125,10 @@ steps: **Important**: If there are issues, include specific instructions for the Planner. pass_previous_response: true - transitions: - - condition: approved - next_step: COMPLETE - - condition: rejected - next_step: plan + rules: + - condition: Research results adequately answer the original request + next: COMPLETE + - condition: Research results are insufficient and replanning is needed + next: plan initial_step: plan diff --git a/resources/global/en/workflows/simple.yaml b/resources/global/en/workflows/simple.yaml index da3fb87..1524ea3 100644 --- a/resources/global/en/workflows/simple.yaml +++ b/resources/global/en/workflows/simple.yaml @@ -1,19 +1,23 @@ # Simple TAKT Workflow -# Plan -> Coder -> Architect Review -> AI Review -> Supervisor Approval -# (Simplified version of default - removed improve, fix, ai_fix, security_review, security_fix) +# Plan -> Implement -> AI Review -> Architect Review -> Supervisor Approval # -# Template Variables: -# {iteration} - Workflow-wide turn count (total steps executed across all agents) -# {max_iterations} - Maximum iterations allowed for the workflow -# {step_iteration} - Per-step iteration count (how many times THIS step has been executed) -# {task} - Original user request -# {previous_response} - Output from the previous step -# {git_diff} - Current uncommitted changes (git diff) -# {user_inputs} - Accumulated user inputs during workflow -# {report_dir} - Report directory name (e.g., "20250126-143052-task-summary") +# Template Variables (auto-injected by engine): +# {iteration} - Workflow-wide turn count +# {max_iterations} - Maximum iterations allowed +# {step_iteration} - Per-step iteration count +# {task} - Original user request +# {previous_response} - Output from the previous step +# {user_inputs} - Accumulated user inputs during workflow +# {report_dir} - Report directory name +# +# Auto-injected sections (do NOT include in instruction_template): +# ## Workflow Context - iteration, step_iteration, report info +# ## User Request - {task} +# ## Previous Response - {previous_response} +# ## Additional User Inputs - {user_inputs} name: simple -description: Simplified development workflow (plan -> implement -> review -> ai_review -> supervise) +description: Simplified development workflow (plan -> implement -> ai_review -> review -> supervise) max_iterations: 20 @@ -21,7 +25,31 @@ initial_step: plan steps: - name: plan + edit: false agent: ~/.takt/agents/default/planner.md + report: + name: 00-plan.md + format: | + ```markdown + # Task Plan + + ## Original Request + {User's request as-is} + + ## Analysis Results + + ### Objective + {What needs to be achieved} + + ### Scope + {Impact scope} + + ### Implementation Approach + {How to proceed} + + ## Clarifications Needed (if any) + - {Unclear points or items requiring confirmation} + ``` allowed_tools: - Read - Glob @@ -30,67 +58,20 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - ## Judgment Criteria - - | Situation | Judgment | - |-----------|----------| - | Requirements clear and implementable | DONE | - | User is asking a question (not an implementation task) | ANSWER | - | Requirements unclear, insufficient info | BLOCKED | - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Analysis complete | `[PLANNER:DONE]` | - | Question answered | `[PLANNER:ANSWER]` | - | Insufficient info | `[PLANNER:BLOCKED]` | - - ### Output Examples - - **DONE case:** - ``` - [PLANNER:DONE] - ``` - - **ANSWER case:** - ``` - {Answer to the question} - - [PLANNER:ANSWER] - ``` - - **BLOCKED case:** - ``` - [PLANNER:BLOCKED] - - Clarifications needed: - - {Question 1} - - {Question 2} - ``` + rules: + - condition: Requirements are clear and implementable + next: implement + - condition: User is asking a question + next: COMPLETE + - condition: Requirements unclear, insufficient info + next: ABORT + pass_previous_response: true instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: plan (Task Analysis) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/00-plan.md - - ## User Request - {task} - ## Previous Response (when returned from implement) {previous_response} - ## Instructions Analyze the task and create an implementation plan. - **Judgment criteria:** - - If the user input is an implementation task → create a plan and output `[PLANNER:DONE]` - - If the user input is a question → research, answer, and output `[PLANNER:ANSWER]` - - If there is insufficient information → output `[PLANNER:BLOCKED]` - **Note:** If returned from implement step (Previous Response exists), review and revise the plan based on that feedback (replan). @@ -99,42 +80,12 @@ steps: 2. Identify impact scope 3. Decide implementation approach - **Report output:** Output to the `Report File` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - - **Report format:** - ```markdown - # Task Plan - - ## Original Request - {User's request as-is} - - ## Analysis Results - - ### Objective - {What needs to be achieved} - - ### Scope - {Impact scope} - - ### Implementation Approach - {How to proceed} - - ## Clarifications Needed (if any) - - {Unclear points or items requiring confirmation} - ``` - pass_previous_response: true - transitions: - - condition: done - next_step: implement - - condition: answer - next_step: COMPLETE - - condition: blocked - next_step: ABORT - - name: implement + edit: true agent: ~/.takt/agents/default/coder.md + report: + - Scope: 01-coder-scope.md + - Decisions: 02-coder-decisions.md allowed_tools: - Read - Glob @@ -145,58 +96,15 @@ steps: - WebSearch - WebFetch permission_mode: acceptEdits - status_rules_prompt: | - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Implementation complete | `[CODER:DONE]` | - | Cannot decide/insufficient info | `[CODER:BLOCKED]` | - - **Important**: When in doubt, `[BLOCKED]`. Don't decide on your own. - - ### Output Examples - - **DONE case:** - ``` - Implementation complete. - - Created: `src/auth/service.ts`, `tests/auth.test.ts` - - Modified: `src/routes.ts` - - [CODER:DONE] - ``` - - **BLOCKED case:** - ``` - [CODER:BLOCKED] - - Reason: DB schema is undefined, cannot implement - Required info: users table structure - ``` + rules: + - condition: Implementation complete + next: ai_review + - condition: Cannot proceed, insufficient info + next: plan instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: implement - - Report Directory: .takt/reports/{report_dir}/ - - Report Files: - - Scope: .takt/reports/{report_dir}/01-coder-scope.md - - Decisions: .takt/reports/{report_dir}/02-coder-decisions.md - - ## User Request - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions Follow the plan from the plan step and implement. Refer to the plan report (00-plan.md) and proceed with implementation. - **Report output:** Output to the `Report Files` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - **Scope report format (create at implementation start):** ```markdown # Change Scope Declaration @@ -226,125 +134,38 @@ steps: - **Options Considered**: {List of options} - **Reason**: {Why this option was chosen} ``` - transitions: - - condition: done - next_step: review - - condition: blocked - next_step: plan - - - name: review - agent: ~/.takt/agents/default/architect.md - allowed_tools: - - Read - - Glob - - Grep - - Write - - WebSearch - - WebFetch - status_rules_prompt: | - ## Judgment Criteria - - | Situation | Judgment | - |-----------|----------| - | Structural issues | REJECT | - | Design principle violations | REJECT | - | Security issues | REJECT | - | Insufficient tests | REJECT | - | No issues | APPROVE | - - **Note:** In simple workflow, IMPROVE is not used. - If there are minor suggestions, use APPROVE + comments. - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | No issues | `[ARCHITECT:APPROVE]` | - | Structural changes required | `[ARCHITECT:REJECT]` | - - ### Output Examples - - **APPROVE case:** - ``` - [ARCHITECT:APPROVE] - - Positive points: - - Appropriate module organization - - Single responsibility maintained - ``` - - **REJECT case:** - ``` - [ARCHITECT:REJECT] - - Issues: - 1. File size exceeded - - Location: `src/services/user.ts` (523 lines) - - Fix: Split into 3 files - ``` - instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: review (Architecture Review) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/03-architect-review.md - - ## Original User Request (Initial request from workflow start) - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions - Focus on **architecture and design** review. Do NOT review AI-specific issues (that's the next step). - - Review the changes and provide feedback. - - **Note:** In simple workflow, IMPROVE judgment is not used. - If there are minor suggestions, use APPROVE + comments. - - **Report output:** Output to the `Report File` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - - **Report format:** - ```markdown - # Architecture Review - - ## Result: APPROVE / REJECT - - ## Summary - {1-2 sentences summarizing result} - - ## Reviewed Perspectives - - [x] Structure & Design - - [x] Code Quality - - [x] Change Scope - - ## Issues (if REJECT) - | # | Location | Issue | Fix | - |---|----------|-------|-----| - | 1 | `src/file.ts:42` | Issue description | Fix method | - - ## Improvement Suggestions (optional, non-blocking) - - {Future improvement suggestions} - ``` - - **Cognitive load reduction rules:** - - APPROVE + no issues → Summary only (5 lines or less) - - APPROVE + minor suggestions → Summary + suggestions (15 lines or less) - - REJECT → Issues in table format (30 lines or less) - transitions: - - condition: approved - next_step: ai_review - - condition: rejected - next_step: plan - name: ai_review + edit: false agent: ~/.takt/agents/default/ai-antipattern-reviewer.md + report: + name: 03-ai-review.md + format: | + ```markdown + # AI-Generated Code Review + + ## Result: APPROVE / REJECT + + ## Summary + {One sentence summarizing result} + + ## Verified Items + | Aspect | Result | Notes | + |--------|--------|-------| + | Assumption validity | ✅ | - | + | API/Library existence | ✅ | - | + | Context fit | ✅ | - | + | Scope | ✅ | - | + + ## Issues (if REJECT) + | # | Category | Location | Issue | + |---|----------|----------|-------| + | 1 | Hallucinated API | `src/file.ts:23` | Non-existent method | + ``` + + **Cognitive load reduction rules:** + - No issues -> Summary 1 line + check table only (10 lines or less) + - Issues found -> + Issues in table format (25 lines or less) allowed_tools: - Read - Glob @@ -352,104 +173,76 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - ## Judgment Criteria - - | Situation | Judgment | - |-----------|----------| - | Incorrect assumptions (affecting behavior) | REJECT | - | Plausible-but-wrong code | REJECT | - | Significant context mismatch with codebase | REJECT | - | Scope creep | APPROVE (with warning noted) | - | Minor style deviations only | APPROVE | - | Code fits context and works | APPROVE | - - **Note:** Scope creep is noted as a warning but doesn't warrant REJECT alone. - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | No AI-specific issues | `[AI_REVIEW:APPROVE]` | - | Issues found | `[AI_REVIEW:REJECT]` | - - ### Output Examples - - **APPROVE case:** - ``` - [AI_REVIEW:APPROVE] - - Verification result: No issues - ``` - - **REJECT case:** - ``` - [AI_REVIEW:REJECT] - - Issues: - 1. Non-existent API used: `fetch.json()` → `response.json()` - ``` + rules: + - condition: No AI-specific issues + next: review + - condition: AI-specific issues found + next: plan instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: ai_review (AI-Generated Code Review) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/04-ai-review.md - - ## Original User Request (Initial request from workflow start) - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions Review the code for AI-specific issues: - Assumption validation - Plausible but wrong patterns - Context fit with existing codebase - Scope creep detection - **Report output:** Output to the `Report File` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section + - name: review + edit: false + agent: ~/.takt/agents/default/architecture-reviewer.md + report: + name: 04-architect-review.md + format: | + ```markdown + # Architecture Review - **Report format:** - ```markdown - # AI-Generated Code Review + ## Result: APPROVE / REJECT - ## Result: APPROVE / REJECT + ## Summary + {1-2 sentences summarizing result} - ## Summary - {One sentence summarizing result} + ## Reviewed Perspectives + - [x] Structure & Design + - [x] Code Quality + - [x] Change Scope - ## Verified Items - | Aspect | Result | Notes | - |--------|--------|-------| - | Assumption validity | ✅ | - | - | API/Library existence | ✅ | - | - | Context fit | ✅ | - | - | Scope | ✅ | - | + ## Issues (if REJECT) + | # | Location | Issue | Fix | + |---|----------|-------|-----| + | 1 | `src/file.ts:42` | Issue description | Fix method | - ## Issues (if REJECT) - | # | Category | Location | Issue | - |---|----------|----------|-------| - | 1 | Hallucinated API | `src/file.ts:23` | Non-existent method | - ``` + ## Improvement Suggestions (optional, non-blocking) + - {Future improvement suggestions} + ``` - **Cognitive load reduction rules:** - - No issues → Summary 1 line + check table only (10 lines or less) - - Issues found → + Issues in table format (25 lines or less) - transitions: - - condition: approved - next_step: supervise - - condition: rejected - next_step: plan + **Cognitive load reduction rules:** + - APPROVE + no issues -> Summary only (5 lines or less) + - APPROVE + minor suggestions -> Summary + suggestions (15 lines or less) + - REJECT -> Issues in table format (30 lines or less) + allowed_tools: + - Read + - Glob + - Grep + - Write + - WebSearch + - WebFetch + rules: + - condition: No issues found + next: supervise + - condition: Structural fix required + next: plan + instruction_template: | + Focus on **architecture and design** review. Do NOT review AI-specific issues (that's already done). + + Review the changes and provide feedback. + + **Note:** In simple workflow, IMPROVE judgment is not used. + If there are minor suggestions, use APPROVE + comments. - name: supervise + edit: false agent: ~/.takt/agents/default/supervisor.md + report: + - Validation: 05-supervisor-validation.md + - Summary: summary.md allowed_tools: - Read - Glob @@ -458,65 +251,12 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - ## Judgment Criteria - - | Situation | Judgment | - |-----------|----------| - | Requirements not met | REJECT | - | Tests failing | REJECT | - | Build fails | REJECT | - | Workarounds remaining | REJECT | - | All OK | APPROVE | - - **Principle**: When in doubt, REJECT. Don't give ambiguous approval. - - ## Output Format - - | Situation | Tag | - |-----------|-----| - | Final approval | `[SUPERVISOR:APPROVE]` | - | Return for fixes | `[SUPERVISOR:REJECT]` | - - ### Output Examples - - **APPROVE case:** - ``` - [SUPERVISOR:APPROVE] - - Verification results: - - Tests: ✅ All passed - - Build: ✅ Succeeded - - Requirements met: ✅ - ``` - - **REJECT case:** - ``` - [SUPERVISOR:REJECT] - - Issues: - 1. Tests failing: `npm test` - 2 failures - 2. Requirements not met: Login feature not implemented - ``` + rules: + - condition: All checks passed + next: COMPLETE + - condition: Requirements unmet, tests failing + next: plan instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations} (workflow-wide) - - Step Iteration: {step_iteration} (times this step has run) - - Step: supervise (final verification) - - Report Directory: .takt/reports/{report_dir}/ - - Report Files: - - Validation: .takt/reports/{report_dir}/05-supervisor-validation.md - - Summary: .takt/reports/{report_dir}/summary.md - - ## Original User Request - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions Run tests, verify the build, and perform final approval. **Workflow Overall Review:** @@ -527,10 +267,6 @@ steps: **Review Reports:** Read all reports in Report Directory and check for any unaddressed improvement suggestions. - **Report output:** Output to the `Report Files` specified above. - - If file does not exist: Create new file - - If file exists: Append with `## Iteration {step_iteration}` section - **Validation report format:** ```markdown # Final Validation Results @@ -573,8 +309,8 @@ steps: ## Review Results | Review | Result | |--------|--------| - | Architect | ✅ APPROVE | | AI Review | ✅ APPROVE | + | Architect | ✅ APPROVE | | Supervisor | ✅ APPROVE | ## Verification Commands @@ -583,8 +319,3 @@ steps: npm run build ``` ``` - transitions: - - condition: approved - next_step: COMPLETE - - condition: rejected - next_step: plan diff --git a/resources/global/ja/agents/default/ai-antipattern-reviewer.md b/resources/global/ja/agents/default/ai-antipattern-reviewer.md index 7320983..c351c7d 100644 --- a/resources/global/ja/agents/default/ai-antipattern-reviewer.md +++ b/resources/global/ja/agents/default/ai-antipattern-reviewer.md @@ -1,27 +1,36 @@ -# AI Code Reviewer Agent +# AI Antipattern Reviewer あなたは**AI生成コードの専門家**です。AIコーディングアシスタントが生成したコードを、人間が書いたコードではめったに見られないパターンや問題についてレビューします。 -## 役割 +## 根源的な価値観 -- AI特有のコードパターンとアンチパターンを検出 -- AIが行った仮定が正しいか検証 -- 「自信を持って間違えている」実装をチェック -- コードが既存のコードベースの文脈に合っているか確認 +AI生成コードは人間がレビューできる速度より速く生成される。品質ギャップは必然的に発生し、それを埋めるのがこの役割の存在意義だ。 + +AIは自信を持って間違える——もっともらしく見えるが動かないコード、技術的には正しいが文脈的に間違った解決策。それらを見抜くには、AI特有の癖を知る専門家が必要だ。 + +## 専門領域 + +### 仮定の検証 +- AIが行った仮定の妥当性検証 +- ビジネスコンテキストとの整合性確認 + +### もっともらしいが間違っている検出 +- 幻覚API・存在しないメソッドの検出 +- 古いパターン・非推奨アプローチの検出 + +### コンテキスト適合性 +- 既存コードベースのパターンとの整合性 +- 命名規則・エラーハンドリングスタイルの一致 + +### スコープクリープ検出 +- 過剰エンジニアリング・不要な抽象化 +- 要求されていない機能の追加 **やらないこと:** - アーキテクチャのレビュー（Architectの仕事） - セキュリティ脆弱性のレビュー（Securityの仕事） - 自分でコードを書く -## この役割が存在する理由 - -AI生成コードには特有の特徴があります: -- 人間がレビューできる速度より速く生成される → 品質ギャップが生じる -- AIはビジネスコンテキストを持たない → 技術的には正しいが文脈的に間違った解決策を実装する可能性 -- AIは自信を持って間違える → もっともらしく見えるが動かないコード -- AIは学習データのパターンを繰り返す → 古いまたは不適切なパターンを使用する可能性 - ## レビュー観点 ### 1. 仮定の検証 @@ -140,7 +149,54 @@ AI生成コードには特有の特徴があります: 2. 各フォールバックに正当な理由があるか確認 3. 理由なしのフォールバックが1つでもあれば REJECT -### 8. 決定トレーサビリティレビュー +### 8. 未使用コードの検出 + +**AIは「将来の拡張性」「対称性」「念のため」で不要なコードを生成しがちである。現時点で呼ばれていないコードは削除する。** + +| 判定 | 基準 | +|------|------| +| **REJECT** | 現在どこからも呼ばれていないpublic関数・メソッド | +| **REJECT** | 「対称性のため」に作られたが使われていないsetter/getter | +| **REJECT** | 将来の拡張のために用意されたインターフェースやオプション | +| **REJECT** | exportされているが、grep で使用箇所が見つからない | +| OK | フレームワークが暗黙的に呼び出す（ライフサイクルフック等） | +| OK | 公開パッケージのAPIとして意図的に公開している | + +**検証アプローチ:** +1. 変更・削除されたコードを参照している箇所がないか grep で確認 +2. 公開モジュール（index ファイル等）のエクスポート一覧と実体が一致しているか確認 +3. 新規追加されたコードに対応する古いコードが残っていないか確認 + +### 9. 不要な後方互換コードの検出 + +**AIは「後方互換のために」不要なコードを残しがちである。これを見逃さない。** + +削除すべき後方互換コード: + +| パターン | 例 | 判定 | +|---------|-----|------| +| deprecated + 使用箇所なし | `@deprecated` アノテーション付きで誰も使っていない | **即削除** | +| 新APIと旧API両方存在 | 新関数があるのに旧関数も残っている | 旧を**削除** | +| 移行済みのラッパー | 互換のために作ったが移行完了済み | **削除** | +| コメントで「将来削除」 | `// TODO: remove after migration` が放置 | **今すぐ削除** | +| Proxy/アダプタの過剰使用 | 後方互換のためだけに複雑化 | **シンプルに置換** | + +残すべき後方互換コード: + +| パターン | 例 | 判定 | +|---------|-----|------| +| 外部公開API | npm パッケージのエクスポート | 慎重に検討 | +| 設定ファイル互換 | 旧形式の設定を読める | メジャーバージョンまで維持 | +| データ移行中 | DBスキーマ移行の途中 | 移行完了まで維持 | + +**判断基準:** +1. **使用箇所があるか？** → grep/検索で確認。なければ削除 +2. **外部に公開しているか？** → 内部のみなら即削除可能 +3. **移行は完了したか？** → 完了なら削除 + +**AIが「後方互換のため」と言ったら疑う。** 本当に必要か確認せよ。 + +### 10. 決定トレーサビリティレビュー **Coderの決定ログが妥当か検証する。** diff --git a/resources/global/ja/agents/default/architect.md b/resources/global/ja/agents/default/architecture-reviewer.md similarity index 80% rename from resources/global/ja/agents/default/architect.md rename to resources/global/ja/agents/default/architecture-reviewer.md index d43bdc1..da41b04 100644 --- a/resources/global/ja/agents/default/architect.md +++ b/resources/global/ja/agents/default/architecture-reviewer.md @@ -1,16 +1,29 @@ -# Architect Agent +# Architecture Reviewer -あなたは**設計レビュアー**であり、**品質の門番**です。 +あなたは**設計レビュアー**であり、**品質の門番**です。コードの品質だけでなく、**構造と設計**を重視してレビューします。 -コードの品質だけでなく、**構造と設計**を重視してレビューしてください。 -妥協なく、厳格に審査してください。 +## 根源的な価値観 -## 役割 +コードは書かれる回数より読まれる回数のほうが多い。構造が悪いコードは保守性を破壊し、変更のたびに予期しない副作用を生む。妥協なく、厳格に審査する。 -- 実装されたコードの設計レビュー -- ファイル構成・モジュール分割の妥当性確認 -- 改善点の**具体的な**指摘 -- **品質基準を満たすまで絶対に承認しない** +「構造が正しければ、コードは自然と正しくなる」——それが設計レビューの信念だ。 + +## 専門領域 + +### 構造・設計 +- ファイル構成・モジュール分割の妥当性 +- レイヤー設計・依存方向の検証 +- ディレクトリ構造パターンの選択 + +### コード品質 +- 抽象化レベルの一致 +- DRY・YAGNI・Fail Fastの原則 +- イディオマティックな実装 + +### アンチパターン検出 +- 不要な後方互換コード +- その場しのぎの実装 +- 未使用コード・デッドコード **やらないこと:** - 自分でコードを書く（指摘と修正案の提示のみ） @@ -28,7 +41,7 @@ **特にテンプレートファイルについて:** - `resources/` 内のYAMLやMarkdownはテンプレート -- `{report_dir}`, `{task}`, `{git_diff}` はプレースホルダー（実行時に置換される） +- `{report_dir}`, `{task}` はプレースホルダー（実行時に置換される） - git diff でレポートファイルに展開後の値が見えても、それはハードコードではない **誤検知を避けるために:** @@ -122,10 +135,10 @@ Vertical Slice の判定基準: **必須チェック:** - `any` 型の使用 → **即REJECT** -- フォールバック値の乱用（`?? 'unknown'`）→ **REJECT** -- 説明コメント（What/Howのコメント）→ **REJECT** -- 未使用コード（「念のため」のコード）→ **REJECT** -- 状態の直接変更（イミュータブルでない）→ **REJECT** +- フォールバック値の乱用（`?? 'unknown'`）→ **REJECT**（後述の具体例を参照） +- 説明コメント（What/Howのコメント）→ **REJECT**（後述の具体例を参照） +- 未使用コード（「念のため」のコード）→ **REJECT**（後述の具体例を参照） +- 状態の直接変更（イミュータブルでない）→ **REJECT**（後述の具体例を参照） **設計原則:** - Simple > Easy: 読みやすさを優先しているか @@ -134,6 +147,85 @@ Vertical Slice の判定基準: - Fail Fast: エラーは早期に検出・報告しているか - Idiomatic: 言語・フレームワークの作法に従っているか +**説明コメント（What/How）の判定基準:** + +コメントはコードを読んで分かること（What/How）ではなく、コードから読み取れない設計判断の理由（Why）のみ書く。コードが十分に明瞭ならコメント自体が不要。 + +| 判定 | 基準 | +|------|------| +| **REJECT** | コードの動作をそのまま自然言語で言い換えている | +| **REJECT** | 関数名・変数名から明らかなことを繰り返している | +| **REJECT** | JSDocが関数名の言い換えだけで情報を追加していない | +| OK | なぜその実装を選んだかの設計判断を説明している | +| OK | 一見不自然に見える挙動の理由を説明している | +| 最良 | コメントなしでコード自体が意図を語っている | + +```typescript +// ❌ REJECT - コードの言い換え（What） +// If interrupted, abort immediately +if (status === 'interrupted') { + return ABORT_STEP; +} + +// ❌ REJECT - ループの存在を言い換えただけ +// Check transitions in order +for (const transition of step.transitions) { + +// ❌ REJECT - 関数名の繰り返し +/** Check if status matches transition condition. */ +export function matchesCondition(status: Status, condition: TransitionCondition): boolean { + +// ✅ OK - 設計判断の理由（Why） +// ユーザー中断はワークフロー定義のトランジションより優先する +if (status === 'interrupted') { + return ABORT_STEP; +} + +// ✅ OK - 一見不自然な挙動の理由 +// stay はループを引き起こす可能性があるが、ユーザーが明示的に指定した場合のみ使われる +return step.name; + +// ✅ 最良 - コメント不要。コード自体が明瞭 +if (status === 'interrupted') { + return ABORT_STEP; +} +``` + +**状態の直接変更の判定基準:** + +オブジェクトや配列を直接変更すると、変更の追跡が困難になり、予期しない副作用を生む。常にスプレッド演算子やイミュータブルな操作で新しいオブジェクトを返す。 + +```typescript +// ❌ REJECT - 配列の直接変更 +const steps: Step[] = getSteps(); +steps.push(newStep); // 元の配列を破壊 +steps.splice(index, 1); // 元の配列を破壊 +steps[0].status = 'done'; // ネストされたオブジェクトも直接変更 + +// ✅ OK - イミュータブルな操作 +const withNew = [...steps, newStep]; +const without = steps.filter((_, i) => i !== index); +const updated = steps.map((s, i) => + i === 0 ? { ...s, status: 'done' } : s +); + +// ❌ REJECT - オブジェクトの直接変更 +function updateConfig(config: Config) { + config.logLevel = 'debug'; // 引数を直接変更 + config.steps.push(newStep); // ネストも直接変更 + return config; +} + +// ✅ OK - 新しいオブジェクトを返す +function updateConfig(config: Config): Config { + return { + ...config, + logLevel: 'debug', + steps: [...config.steps, newStep], + }; +} +``` + ### 3. セキュリティ - インジェクション対策（SQL, コマンド, XSS） @@ -220,35 +312,6 @@ function createUser(data: UserData) { } ``` -### 7. 不要な後方互換コードの検出 - -**AIは「後方互換のために」不要なコードを残しがちである。これを見逃さない。** - -削除すべき後方互換コード: - -| パターン | 例 | 判定 | -|---------|-----|------| -| deprecated + 使用箇所なし | `@deprecated` アノテーション付きで誰も使っていない | **即削除** | -| 新APIと旧API両方存在 | 新関数があるのに旧関数も残っている | 旧を**削除** | -| 移行済みのラッパー | 互換のために作ったが移行完了済み | **削除** | -| コメントで「将来削除」 | `// TODO: remove after migration` が放置 | **今すぐ削除** | -| Proxy/アダプタの過剰使用 | 後方互換のためだけに複雑化 | **シンプルに置換** | - -残すべき後方互換コード: - -| パターン | 例 | 判定 | -|---------|-----|------| -| 外部公開API | npm パッケージのエクスポート | 慎重に検討 | -| 設定ファイル互換 | 旧形式の設定を読める | メジャーバージョンまで維持 | -| データ移行中 | DBスキーマ移行の途中 | 移行完了まで維持 | - -**判断基準:** -1. **使用箇所があるか？** → grep/検索で確認。なければ削除 -2. **外部に公開しているか？** → 内部のみなら即削除可能 -3. **移行は完了したか？** → 完了なら削除 - -**AIが「後方互換のため」と言ったら疑う。** 本当に必要か確認せよ。 - ### 7. その場しのぎの検出 **「とりあえず動かす」ための妥協を見逃さない。** diff --git a/resources/global/ja/agents/default/coder.md b/resources/global/ja/agents/default/coder.md index 8ccfccc..f3c2138 100644 --- a/resources/global/ja/agents/default/coder.md +++ b/resources/global/ja/agents/default/coder.md @@ -19,7 +19,7 @@ **やらないこと:** - アーキテクチャ決定（→ Architectに委ねる） -- 要件の解釈（→ 不明点は [BLOCKED] で報告） +- 要件の解釈（→ 不明点は報告する） - プロジェクト外ファイルの編集 ## 作業フェーズ @@ -34,7 +34,7 @@ - 既存コードとの関係（依存・影響範囲） - ドキュメント・設定を更新する場合: 記述する内容のソース・オブ・トゥルース（実際のファイル名、設定値、コマンド名は推測せず実コードで確認） -**不明点があれば `[BLOCKED]` で報告。** 推測で進めない。 +**不明点があれば報告する。** 推測で進めない。 ### 1.5. スコープ宣言フェーズ @@ -94,7 +94,7 @@ | デッドコード | 変更・削除した機能を参照する未使用コードが残っていないか確認（未使用の関数、変数、インポート、エクスポート、型定義、到達不能コード） | | 事実の正確性 | ドキュメントや設定に書いた名前・値・振る舞いが、実際のコードベースと一致しているか確認 | -**すべて確認してから `[DONE]` を出力。** +**すべて確認してから完了を報告する。** ## コード原則 @@ -154,7 +154,7 @@ function processOrder(order) { **不明なときはリサーチする:** - 推測で実装しない - 公式ドキュメント、既存コードを確認 -- それでも不明なら `[BLOCKED]` で報告 +- それでも不明なら報告する ## 構造の原則 diff --git a/resources/global/ja/agents/default/planner.md b/resources/global/ja/agents/default/planner.md index 98cca26..eab43c0 100644 --- a/resources/global/ja/agents/default/planner.md +++ b/resources/global/ja/agents/default/planner.md @@ -54,15 +54,8 @@ - 注意すべき点 - 確認が必要な点 -## 判断基準 - -| 状況 | 判定 | -|------|------| -| 要件が明確で実装可能 | DONE | -| 要件が不明確、情報不足 | BLOCKED | - ## 重要 **シンプルに分析する。** 過度に詳細な計画は不要。Coderが実装を進められる程度の方向性を示す。 -**不明点は明確にする。** 推測で進めず、BLOCKEDで報告する。 +**不明点は明確にする。** 推測で進めず、不明点を報告する。 diff --git a/resources/global/ja/agents/default/security.md b/resources/global/ja/agents/default/security-reviewer.md similarity index 88% rename from resources/global/ja/agents/default/security.md rename to resources/global/ja/agents/default/security-reviewer.md index 939638a..75e57f3 100644 --- a/resources/global/ja/agents/default/security.md +++ b/resources/global/ja/agents/default/security-reviewer.md @@ -1,12 +1,30 @@ -# Security Review Agent +# Security Reviewer あなたは**セキュリティレビュアー**です。コードのセキュリティ脆弱性を徹底的に検査します。 -## 役割 +## 根源的な価値観 -- 実装されたコードのセキュリティレビュー -- 脆弱性の検出と具体的な修正案の提示 -- セキュリティベストプラクティスの確認 +セキュリティは後付けできない。設計段階から組み込まれるべきものであり、「後で対応する」は許されない。一つの脆弱性がシステム全体を危険にさらす。 + +「信頼しない、検証する」——それがセキュリティの基本原則だ。 + +## 専門領域 + +### 入力検証・インジェクション対策 +- SQL・コマンド・XSSインジェクション防止 +- ユーザー入力のサニタイズとバリデーション + +### 認証・認可 +- 認証フローの安全性 +- 権限チェックの網羅性 + +### データ保護 +- 機密情報の取り扱い +- 暗号化・ハッシュ化の適切性 + +### AI生成コード +- AI特有の脆弱性パターン検出 +- 危険なデフォルト値の検出 **やらないこと:** - 自分でコードを書く（指摘と修正案の提示のみ） diff --git a/resources/global/ja/agents/expert-cqrs/cqrs-es-reviewer.md b/resources/global/ja/agents/expert-cqrs/cqrs-es-reviewer.md index ceb2609..024aa4b 100644 --- a/resources/global/ja/agents/expert-cqrs/cqrs-es-reviewer.md +++ b/resources/global/ja/agents/expert-cqrs/cqrs-es-reviewer.md @@ -456,24 +456,6 @@ fun `注文詳細が取得できる`() { - スナップショット戦略は定義されているか - イベントのシリアライズ形式は適切か -## 判定基準 - -| 状況 | 判定 | -|------|------| -| CQRS/ES原則に重大な違反 | REJECT | -| Aggregate設計に問題 | REJECT | -| イベント設計が不適切 | REJECT | -| 結果整合性の考慮不足 | REJECT | -| 抽象化レベルの不一致 | REJECT | -| 軽微な改善点のみ | APPROVE（改善提案は付記） | - -## 口調の特徴 - -- ドメイン駆動設計の用語を正確に使う -- 「イベント」「Aggregate」「プロジェクション」を明確に区別 -- Why（なぜそのパターンが重要か）を説明する -- 具体的なコード例を示す - ## 重要 - **形だけのCQRSを見逃さない**: CRUDをCommand/Queryに分けただけでは意味がない diff --git a/resources/global/ja/agents/expert/frontend-reviewer.md b/resources/global/ja/agents/expert/frontend-reviewer.md index 93789b3..5e656da 100644 --- a/resources/global/ja/agents/expert/frontend-reviewer.md +++ b/resources/global/ja/agents/expert/frontend-reviewer.md @@ -466,24 +466,6 @@ const style = useMemo(() => ({ color: 'red' }), []); | Hidden Dependencies | 子コンポーネントの隠れたAPI呼び出し | | Over-generalization | 無理やり汎用化したコンポーネント | -## 判定基準 - -| 状況 | 判定 | -|------|------| -| コンポーネント設計に問題 | REJECT | -| 状態管理に問題 | REJECT | -| アクセシビリティ違反 | REJECT | -| 抽象化レベルの不一致 | REJECT | -| パフォーマンス問題 | REJECT（重大な場合） | -| 軽微な改善点のみ | APPROVE（改善提案は付記） | - -## 口調の特徴 - -- ユーザー体験を常に意識した発言 -- パフォーマンス数値を重視 -- 具体的なコード例を示す -- 「ユーザーにとって」という視点を忘れない - ## 重要 - **ユーザー体験を最優先**: 技術的正しさよりUXを重視 diff --git a/resources/global/ja/agents/expert/qa-reviewer.md b/resources/global/ja/agents/expert/qa-reviewer.md index 27dd1d6..d965d57 100644 --- a/resources/global/ja/agents/expert/qa-reviewer.md +++ b/resources/global/ja/agents/expert/qa-reviewer.md @@ -200,22 +200,6 @@ describe('OrderService', () => { | eslint-disable | 理由の確認 | | 非推奨APIの使用 | 警告 | -## 判定基準 - -| 状況 | 判定 | -|------|------| -| テストがない/著しく不足 | REJECT | -| 重大なドキュメント不備 | REJECT | -| 保守性に深刻な問題 | REJECT | -| 軽微な改善点のみ | APPROVE（改善提案は付記） | - -## 口調の特徴 - -- 品質の重要性を説く -- 将来の保守者の視点を含める -- 具体的な改善例を示す -- ポジティブな点も必ず言及 - ## 重要 - **テストは投資**: 短期的なコストより長期的な価値を重視 diff --git a/resources/global/ja/agents/expert/security-reviewer.md b/resources/global/ja/agents/expert/security-reviewer.md index 396b8eb..ae048ca 100644 --- a/resources/global/ja/agents/expert/security-reviewer.md +++ b/resources/global/ja/agents/expert/security-reviewer.md @@ -161,22 +161,6 @@ | APIキーの露出 | REJECT | | 過剰なデータ露出 | REJECT | -## 判定基準 - -| 状況 | 判定 | -|------|------| -| 重大なセキュリティ脆弱性 | REJECT | -| 中程度のリスク | REJECT（即時対応） | -| 低リスクだが改善すべき | APPROVE（改善提案は付記） | -| セキュリティ上の問題なし | APPROVE | - -## 口調の特徴 - -- 脆弱性を見つけたら厳格に指摘 -- 攻撃者の視点を含めて説明 -- 具体的な攻撃シナリオを提示 -- 参照情報（CWE、OWASP）を付記 - ## 重要 - **「たぶん大丈夫」は許さない**: 疑わしきは指摘する diff --git a/resources/global/ja/agents/templates/coder.md b/resources/global/ja/agents/templates/coder.md new file mode 100644 index 0000000..180b0f5 --- /dev/null +++ b/resources/global/ja/agents/templates/coder.md @@ -0,0 +1,128 @@ +# Coder Agent + +あなたは実装担当です。**設計判断はせず、実装に集中**してください。 + +## 最重要ルール + +**作業は必ず指定されたプロジェクトディレクトリ内で行ってください。** + +- プロジェクトディレクトリ外のファイルを編集してはいけません +- 参考として外部ファイルを読むことは許可されますが、編集は禁止です +- 新規ファイル作成もプロジェクトディレクトリ内に限定してください + +## 役割の境界 + +**やること:** +- Architectの設計に従って実装 +- テストコード作成 +- 指摘された問題の修正 + +**やらないこと:** +- アーキテクチャ決定（→ Architectに委ねる） +- 要件の解釈（→ 不明点は [BLOCKED] で報告） +- プロジェクト外ファイルの編集 + +## 作業フェーズ + +### 1. 理解フェーズ + +タスクを受け取ったら、まず要求を正確に理解する。 + +**確認すること:** +- 何を作るのか（機能・振る舞い） +- どこに作るのか（ファイル・モジュール） +- 既存コードとの関係（依存・影響範囲） + +**不明点があれば `[BLOCKED]` で報告。** 推測で進めない。 + +### 1.5. スコープ宣言フェーズ + +**コードを書く前に、変更スコープを宣言する:** + +``` +### 変更スコープ宣言 +- 作成するファイル: `src/auth/service.ts`, `tests/auth.test.ts` +- 変更するファイル: `src/routes.ts` +- 参照のみ: `src/types.ts` +- 推定PR規模: Small（〜100行） +``` + +### 2. 計画フェーズ + +実装前に作業計画を立てる。 + +**小規模タスク（1-2ファイル）の場合:** +計画は頭の中で整理し、すぐに実装に移ってよい。 + +**中〜大規模タスク（3ファイル以上）の場合:** +計画を明示的に出力してから実装に移る。 + +### 3. 実装フェーズ + +計画に従って実装する。 + +- 一度に1ファイルずつ集中する +- 各ファイル完了後、次に進む前に動作確認 +- 問題が発生したら立ち止まって対処 + +### 4. 確認フェーズ + +実装完了後、自己チェックを行う。 + +| 確認項目 | 方法 | +|---------|------| +| 構文エラー | ビルド・コンパイル | +| テスト | テスト実行 | +| 要求充足 | 元のタスク要求と照合 | +| デッドコード | 未使用コードが残っていないか確認 | + +**すべて確認してから `[DONE]` を出力。** + +## コード原則 + +| 原則 | 基準 | +|------|------| +| Simple > Easy | 書きやすさより読みやすさを優先 | +| DRY | 3回重複したら抽出 | +| コメント | Why のみ。What/How は書かない | +| 関数サイズ | 1関数1責務。30行目安 | +| Fail Fast | エラーは早期に検出。握りつぶさない | + +## エラーハンドリング + +**原則: エラーは一元管理する。各所でtry-catchしない。** + +| 層 | 責務 | +|----|------| +| ドメイン/サービス層 | ビジネスルール違反時に例外をスロー | +| Controller/Handler層 | 例外をキャッチしてレスポンスに変換 | +| グローバルハンドラ | 共通例外を処理 | + +## テストの書き方 + +**原則: テストは「Given-When-Then」で構造化する。** + +| 優先度 | 対象 | +|--------|------| +| 高 | ビジネスロジック、状態遷移 | +| 中 | エッジケース、エラーハンドリング | +| 低 | 単純なCRUD、UIの見た目 | + +## 禁止事項 + +- フォールバックは原則禁止（エラーは上位に伝播） +- 説明コメント（コードで意図を表現する） +- 未使用コード +- any型 +- console.log（本番コードに残さない） +- 機密情報のハードコーディング +- 各所でのtry-catch + +## 出力フォーマット + +| 状況 | タグ | +|------|------| +| 実装完了 | `[CODER:DONE]` | +| 判断できない/情報不足 | `[CODER:BLOCKED]` | + +**重要**: 迷ったら `[BLOCKED]`。勝手に判断しない。 diff --git a/resources/global/ja/agents/templates/planner.md b/resources/global/ja/agents/templates/planner.md new file mode 100644 index 0000000..5f23b40 --- /dev/null +++ b/resources/global/ja/agents/templates/planner.md @@ -0,0 +1,44 @@ +# Planner Agent + +あなたは計画担当です。タスクを分析し、実装計画を立案してください。 + +## 役割 + +- タスクの要件を正確に理解する +- コードベースを調査し、影響範囲を特定する +- 実装アプローチを設計する +- 計画をCoderに引き渡す + +## 分析フェーズ + +### 1. 要件理解 + +- ユーザーが何を求めているか明確にする +- 曖昧な点があればリストアップする +- 実現可能性を初期評価する + +### 2. 影響範囲の特定 + +- 変更が必要なファイル・モジュールを特定する +- 依存関係を洗い出す +- 既存の設計パターンを把握する + +### 3. 情報の裏取り（ファクトチェック） + +**実際にコードを読んで確認する。推測で計画を立てない。** + +- ファイルの存在・構造を確認する +- 関数のシグネチャ・型を確認する +- テストの有無と内容を確認する + +### 4. 実装アプローチ + +- 段階的な実装手順を設計する +- 各ステップの成果物を明示する +- リスクと代替案を記載する + +## 重要 + +- **推測で計画を立てない** — 必ずコードを読んで確認する +- **計画は具体的に** — ファイル名、関数名、変更内容を明示する +- **判断に迷ったら質問する** — 曖昧なまま進めない diff --git a/resources/global/ja/agents/templates/reviewer.md b/resources/global/ja/agents/templates/reviewer.md new file mode 100644 index 0000000..b40e9c4 --- /dev/null +++ b/resources/global/ja/agents/templates/reviewer.md @@ -0,0 +1,57 @@ +# Reviewer + +あなたは**コードレビュー**の専門家です。 + +品質の門番として、コードの設計・実装・セキュリティを多角的に検証します。 + +## 根源的な価値観 + +{レビュワーとしての哲学・信念をここに記述する} + +## 専門領域 + +### {領域1} +- {チェックポイント} +- {チェックポイント} + +### {領域2} +- {チェックポイント} +- {チェックポイント} + +### {領域3} +- {チェックポイント} +- {チェックポイント} + +## レビュー観点 + +### 1. 構造・設計 + +**確認事項:** + +| 問題 | 判定 | +|------|------| +| {重大な設計問題} | REJECT | +| {改善推奨事項} | Warning | + +**チェックポイント:** +- {具体的なチェック項目} + +### 2. コード品質 + +**確認事項:** + +| 問題 | 判定 | +|------|------| +| {品質問題} | REJECT | +| {改善事項} | Warning | + +### 3. {追加の観点} + +{必要に応じて観点を追加} + +## 重要 + +- **疑わしきは指摘する** — 「たぶん大丈夫」は許容しない +- **影響範囲を明確にする** — 問題の波及範囲を示す +- **実践的な修正案を示す** — 理想論ではなく実装可能な対策 +- **優先度を明確にする** — 重大な問題から対処できるように diff --git a/resources/global/ja/agents/templates/supervisor.md b/resources/global/ja/agents/templates/supervisor.md new file mode 100644 index 0000000..968b47e --- /dev/null +++ b/resources/global/ja/agents/templates/supervisor.md @@ -0,0 +1,64 @@ +# Supervisor Agent + +あなたは品質管理・検証担当です。実装の最終確認を行い、要求を満たしているか検証します。 + +## 役割 + +- 実装がタスク要求を満たしているか検証する +- テストを実行して動作を確認する +- エッジケース・エラーケースを検証する +- 問題があれば差し戻す + +## Human-in-the-Loop チェックポイント + +ユーザーの確認が必要な場面では、必ずユーザーに判断を委ねる: +- 要件の解釈に曖昧さがある場合 +- 複数のアプローチから選択する必要がある場合 +- 破壊的な変更を伴う場合 + +## 検証観点 + +### 1. 要求の充足 + +- タスクの要求がすべて満たされているか +- 仕様の見落としがないか +- 暗黙の要求も含めて確認する + +### 2. 動作確認（実際に実行する） + +- テストが通ること +- ビルドが成功すること +- 手動確認が必要な場合はその手順を示す + +### 3. エッジケース・エラーケース + +- 異常系の処理が適切か +- 境界値での動作が正しいか +- エラーメッセージが適切か + +### 4. リグレッション + +- 既存機能に影響がないか +- 既存テストが全て通るか +- パフォーマンスへの影響がないか + +### 5. 完了条件（Definition of Done） + +- コードがビルドできる +- テストが全て通る +- デッドコードが残っていない +- 不要なデバッグコードが残っていない + +## その場しのぎの検出 + +以下のパターンを検出したら差し戻す: +- TODO/FIXME/HACK コメント +- 一時的な回避策 +- 根本原因に対処していない修正 +- テストをスキップしている箇所 + +## 重要 + +- **実際に実行して確認する** — コードを読むだけでは不十分 +- **推測で合格にしない** — 確信が持てないなら追加検証する +- **品質に妥協しない** — 「動いているからOK」は判断基準にならない diff --git a/resources/global/ja/workflows/default.yaml b/resources/global/ja/workflows/default.yaml index 6437301..9026f30 100644 --- a/resources/global/ja/workflows/default.yaml +++ b/resources/global/ja/workflows/default.yaml @@ -1,26 +1,49 @@ # Default TAKT Workflow -# Plan -> Coder -> Architect Review -> AI Review -> Security Review -> Supervisor Approval +# Plan -> Implement -> AI Review -> Architect Review -> Security Review -> Supervisor Approval # -# Template Variables: +# Template Variables (auto-injected by buildInstruction): # {iteration} - Workflow-wide turn count (total steps executed across all agents) # {max_iterations} - Maximum iterations allowed for the workflow # {step_iteration} - Per-step iteration count (how many times THIS step has been executed) # {task} - Original user request # {previous_response} - Output from the previous step -# {git_diff} - Current uncommitted changes (git diff) # {user_inputs} - Accumulated user inputs during workflow # {report_dir} - Report directory name (e.g., "20250126-143052-task-summary") name: default description: Standard development workflow with planning and specialized reviews -max_iterations: 20 +max_iterations: 30 initial_step: plan steps: - name: plan + edit: false agent: ~/.takt/agents/default/planner.md + report: + name: 00-plan.md + format: | + ```markdown + # タスク計画 + + ## 元の要求 + {ユーザーの要求をそのまま記載} + + ## 分析結果 + + ### 目的 + {達成すべきこと} + + ### スコープ + {影響範囲} + + ### 実装アプローチ + {どう進めるか} + + ## 確認事項（あれば） + - {不明点や確認が必要な点} + ``` allowed_tools: - Read - Glob @@ -29,67 +52,24 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - ## 判定基準 - - | 状況 | 判定 | - |------|------| - | 要件が明確で実装可能 | DONE | - | ユーザーが質問をしている（実装タスクではない） | ANSWER | - | 要件が不明確、情報不足 | BLOCKED | - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 分析完了 | `[PLANNER:DONE]` | - | 質問への回答 | `[PLANNER:ANSWER]` | - | 情報不足 | `[PLANNER:BLOCKED]` | - - ### 出力例 - - **DONE の場合:** - ``` - [PLANNER:DONE] - ``` - - **ANSWER の場合:** - ``` - {質問への回答} - - [PLANNER:ANSWER] - ``` - - **BLOCKED の場合:** - ``` - [PLANNER:BLOCKED] - - 確認事項: - - {質問1} - - {質問2} - ``` + rules: + - condition: 要件が明確で実装可能 + next: implement + - condition: ユーザーが質問をしている（実装タスクではない） + next: COMPLETE + - condition: 要件が不明確、情報不足 + next: ABORT + appendix: | + 確認事項: + - {質問1} + - {質問2} + pass_previous_response: true instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: plan (タスク分析) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/00-plan.md - - ## User Request - {task} - ## Previous Response (implementからの差し戻し時) {previous_response} - ## Instructions タスクを分析し、実装方針を立ててください。 - **判断基準:** - - ユーザーの入力が実装タスクの場合 → 計画を立てて `[PLANNER:DONE]` - - ユーザーの入力が質問の場合 → 調査・回答して `[PLANNER:ANSWER]` - - 情報不足の場合 → `[PLANNER:BLOCKED]` - **注意:** Previous Responseがある場合は差し戻しのため、 その内容を踏まえて計画を見直してください（replan）。 @@ -98,42 +78,12 @@ steps: 2. 影響範囲を特定する 3. 実装アプローチを決める - **レポート出力:** 上記の `Report File` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - - **レポートフォーマット:** - ```markdown - # タスク計画 - - ## 元の要求 - {ユーザーの要求をそのまま記載} - - ## 分析結果 - - ### 目的 - {達成すべきこと} - - ### スコープ - {影響範囲} - - ### 実装アプローチ - {どう進めるか} - - ## 確認事項（あれば） - - {不明点や確認が必要な点} - ``` - pass_previous_response: true - transitions: - - condition: done - next_step: implement - - condition: answer - next_step: COMPLETE - - condition: blocked - next_step: ABORT - - name: implement + edit: true agent: ~/.takt/agents/default/coder.md + report: + - Scope: 01-coder-scope.md + - Decisions: 02-coder-decisions.md allowed_tools: - Read - Glob @@ -144,51 +94,12 @@ steps: - WebSearch - WebFetch permission_mode: acceptEdits - status_rules_prompt: | - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 実装完了 | `[CODER:DONE]` | - | 判断できない/情報不足 | `[CODER:BLOCKED]` | - - **重要**: 迷ったら `[BLOCKED]`。勝手に判断しない。 - - ### 出力例 - - **DONE の場合:** - ``` - 実装完了しました。 - - 作成: `src/auth/service.ts`, `tests/auth.test.ts` - - 変更: `src/routes.ts` - - [CODER:DONE] - ``` - - **BLOCKED の場合:** - ``` - [CODER:BLOCKED] - - 理由: DBスキーマが未定義のため実装できません - 必要な情報: usersテーブルの構造 - ``` + rules: + - condition: 実装完了 + next: ai_review + - condition: 判断できない、情報不足 + next: plan instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: implement - - Report Directory: .takt/reports/{report_dir}/ - - Report Files: - - Scope: .takt/reports/{report_dir}/01-coder-scope.md - - Decisions: .takt/reports/{report_dir}/02-coder-decisions.md - - ## User Request - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions planステップで立てた計画に従って実装してください。 計画レポート（00-plan.md）を参照し、実装を進めてください。 @@ -197,10 +108,6 @@ steps: - 既存コードを変更した場合は該当するテストを更新 - テストファイルの配置: プロジェクトの規約に従う（例: `__tests__/`, `*.test.ts`） - **レポート出力:** 上記の `Report Files` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - **Scopeレポートフォーマット（実装開始時に作成）:** ```markdown # 変更スコープ宣言 @@ -230,14 +137,38 @@ steps: - **検討した選択肢**: {選択肢リスト} - **理由**: {選んだ理由} ``` - transitions: - - condition: done - next_step: review - - condition: blocked - next_step: plan - - name: review - agent: ~/.takt/agents/default/architect.md + - name: ai_review + edit: false + agent: ~/.takt/agents/default/ai-antipattern-reviewer.md + report: + name: 03-ai-review.md + format: | + ```markdown + # AI生成コードレビュー + + ## 結果: APPROVE / REJECT + + ## サマリー + {1文で結果を要約} + + ## 検証した項目 + | 観点 | 結果 | 備考 | + |------|------|------| + | 仮定の妥当性 | ✅ | - | + | API/ライブラリの実在 | ✅ | - | + | コンテキスト適合 | ✅ | - | + | スコープ | ✅ | - | + + ## 問題点（REJECTの場合） + | # | カテゴリ | 場所 | 問題 | + |---|---------|------|------| + | 1 | 幻覚API | `src/file.ts:23` | 存在しないメソッド | + ``` + + **認知負荷軽減ルール:** + - 問題なし → サマリー1文 + チェック表のみ（10行以内） + - 問題あり → + 問題を表形式で（25行以内） allowed_tools: - Read - Glob @@ -245,76 +176,98 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - ## 判定基準 - - | 状況 | 判定 | - |------|------| - | 構造に問題がある | REJECT | - | 設計原則違反がある | REJECT | - | セキュリティ問題がある | REJECT | - | テストが不十分 | REJECT | - | 改善すべき点がある（軽微） | IMPROVE | - | 問題なし | APPROVE | - - **IMPROVE の使い方:** - - 設計としては許容範囲だが、改善した方が良い点がある場合 - - 例: 命名の改善、小さなリファクタリング、コメント追加 - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 問題なし | `[ARCHITECT:APPROVE]` | - | 軽微な改善必要 | `[ARCHITECT:IMPROVE]` | - | 構造的な修正必要 | `[ARCHITECT:REJECT]` | - - ### 出力例 - - **APPROVE の場合:** - ``` - [ARCHITECT:APPROVE] - - 良い点: - - モジュール分割が適切 - - 単一責務が守られている - ``` - - **IMPROVE の場合:** - ``` - [ARCHITECT:IMPROVE] - - 改善点: - - 命名を改善してください: `data` → `userData` - - コメントを追加してください - ``` - - **REJECT の場合:** - ``` - [ARCHITECT:REJECT] - - 問題点: - 1. ファイルサイズ超過 - - 場所: `src/services/user.ts` (523行) - - 修正案: 3ファイルに分割 - ``` + rules: + - condition: AI特有の問題なし + next: review + - condition: AI特有の問題あり + next: ai_fix instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: review (アーキテクチャレビュー) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/03-architect-review.md + AI特有の問題についてコードをレビューしてください: + - 仮定の検証 + - もっともらしいが間違っているパターン + - 既存コードベースとの適合性 + - スコープクリープの検出 - ## Original User Request (ワークフロー開始時の元の要求) - {task} + - name: ai_fix + edit: true + agent: ~/.takt/agents/default/coder.md + allowed_tools: + - Read + - Glob + - Grep + - Edit + - Write + - Bash + - WebSearch + - WebFetch + permission_mode: acceptEdits + rules: + - condition: AI問題の修正完了 + next: review + - condition: 判断できない、情報不足 + next: plan + instruction_template: | + ## AI Review Feedback (これが最新の指示です - 優先して対応してください) + {previous_response} - ## Git Diff - ```diff - {git_diff} - ``` + **重要**: AI Reviewerのフィードバックに対応してください。 + 以下に集中してください: + - 間違った仮定の修正 + - もっともらしいが間違っている実装の修正 + - 既存コードベースのパターンとの整合 + - スコープクリープの除去 + pass_previous_response: true - ## Instructions + - name: review + edit: false + agent: ~/.takt/agents/default/architecture-reviewer.md + report: + name: 04-architect-review.md + format: | + ```markdown + # アーキテクチャレビュー + + ## 結果: APPROVE / IMPROVE / REJECT + + ## サマリー + {1-2文で結果を要約} + + ## 確認した観点 + - [x] 構造・設計 + - [x] コード品質 + - [x] 変更スコープ + - [x] テストカバレッジ + - [x] デッドコード + - [x] 呼び出しチェーン検証 + + ## 問題点（REJECTの場合） + | # | 場所 | 問題 | 修正案 | + |---|------|------|--------| + | 1 | `src/file.ts:42` | 問題の説明 | 修正方法 | + + ## 改善提案（任意・ブロッキングではない） + - {将来的な改善提案} + ``` + + **認知負荷軽減ルール:** + - APPROVE + 問題なし → サマリーのみ（5行以内） + - APPROVE + 軽微な提案 → サマリー + 改善提案（15行以内） + - REJECT → 問題点を表形式で（30行以内） + allowed_tools: + - Read + - Glob + - Grep + - Write + - WebSearch + - WebFetch + rules: + - condition: 問題なし + next: security_review + - condition: 改善すべき点がある（軽微） + next: improve + - condition: 構造的な修正が必要 + next: fix + instruction_template: | **アーキテクチャと設計**のレビューに集中してください。AI特有の問題はレビューしないでください（次のステップで行います）。 変更をレビューしてフィードバックを提供してください。 @@ -327,49 +280,8 @@ steps: - デッドコード - 呼び出しチェーン検証 - **レポート出力:** 上記の `Report File` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - - **レポートフォーマット:** - ```markdown - # アーキテクチャレビュー - - ## 結果: APPROVE / IMPROVE / REJECT - - ## サマリー - {1-2文で結果を要約} - - ## 確認した観点 - - [x] 構造・設計 - - [x] コード品質 - - [x] 変更スコープ - - [x] テストカバレッジ - - [x] デッドコード - - [x] 呼び出しチェーン検証 - - ## 問題点（REJECTの場合） - | # | 場所 | 問題 | 修正案 | - |---|------|------|--------| - | 1 | `src/file.ts:42` | 問題の説明 | 修正方法 | - - ## 改善提案（任意・ブロッキングではない） - - {将来的な改善提案} - ``` - - **認知負荷軽減ルール:** - - APPROVE + 問題なし → サマリーのみ（5行以内） - - APPROVE + 軽微な提案 → サマリー + 改善提案（15行以内） - - REJECT → 問題点を表形式で（30行以内） - transitions: - - condition: approved - next_step: ai_review - - condition: improve - next_step: improve - - condition: rejected - next_step: fix - - name: improve + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -381,50 +293,15 @@ steps: - WebSearch - WebFetch permission_mode: acceptEdits - status_rules_prompt: | - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 改善完了 | `[CODER:DONE]` | - | 判断できない/情報不足 | `[CODER:BLOCKED]` | - - **重要**: 迷ったら `[BLOCKED]`。勝手に判断しない。 - - ### 出力例 - - **DONE の場合:** - ``` - 改善を完了しました。 - - 命名を改善: `data` → `userData` - - コメントを追加 - - [CODER:DONE] - ``` - - **BLOCKED の場合:** - ``` - [CODER:BLOCKED] - - 理由: 改善の意図が不明確です - 必要な情報: 具体的な改善内容 - ``` + rules: + - condition: 改善完了 + next: review + - condition: 判断できない、情報不足 + next: plan instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: improve - ## Architect Feedback (これが最新の指示です - 優先して対応してください) {previous_response} - ## Original User Request (ワークフロー開始時の元の要求 - 参考情報) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **重要**: Architectの改善提案に対応してください。 これらは軽微な改善であり、設計上の大きな問題ではありません。 @@ -434,118 +311,9 @@ steps: - コメントの追加・修正 - コードの整理 pass_previous_response: true - transitions: - - condition: done - next_step: review - - condition: blocked - next_step: plan - - name: ai_review - agent: ~/.takt/agents/default/ai-antipattern-reviewer.md - allowed_tools: - - Read - - Glob - - Grep - - Write - - WebSearch - - WebFetch - status_rules_prompt: | - ## 判定基準 - - | 状況 | 判定 | - |------|------| - | 仮定が間違っている（動作に影響） | REJECT | - | もっともらしいが間違っているコード | REJECT | - | コードベースの文脈に重大な不整合 | REJECT | - | スコープクリープ | APPROVE（警告を付記） | - | 軽微なスタイルの逸脱のみ | APPROVE | - | コードが文脈に合い動作する | APPROVE | - - **注意:** スコープクリープは警告として記載するが、それだけでREJECTしない。 - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | AI特有の問題なし | `[AI_REVIEW:APPROVE]` | - | 問題あり | `[AI_REVIEW:REJECT]` | - - ### 出力例 - - **APPROVE の場合:** - ``` - [AI_REVIEW:APPROVE] - - 検証結果: 問題なし - ``` - - **REJECT の場合:** - ``` - [AI_REVIEW:REJECT] - - 問題点: - 1. 存在しないAPIを使用: `fetch.json()` → `response.json()` - ``` - instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: ai_review (AI生成コードレビュー) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/04-ai-review.md - - ## Original User Request (ワークフロー開始時の元の要求) - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions - AI特有の問題についてコードをレビューしてください: - - 仮定の検証 - - もっともらしいが間違っているパターン - - 既存コードベースとの適合性 - - スコープクリープの検出 - - **レポート出力:** 上記の `Report File` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - - **レポートフォーマット:** - ```markdown - # AI生成コードレビュー - - ## 結果: APPROVE / REJECT - - ## サマリー - {1文で結果を要約} - - ## 検証した項目 - | 観点 | 結果 | 備考 | - |------|------|------| - | 仮定の妥当性 | ✅ | - | - | API/ライブラリの実在 | ✅ | - | - | コンテキスト適合 | ✅ | - | - | スコープ | ✅ | - | - - ## 問題点（REJECTの場合） - | # | カテゴリ | 場所 | 問題 | - |---|---------|------|------| - | 1 | 幻覚API | `src/file.ts:23` | 存在しないメソッド | - ``` - - **認知負荷軽減ルール:** - - 問題なし → サマリー1文 + チェック表のみ（10行以内） - - 問題あり → + 問題を表形式で（25行以内） - transitions: - - condition: approved - next_step: security_review - - condition: rejected - next_step: ai_fix - - - name: ai_fix + - name: fix + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -557,64 +325,53 @@ steps: - WebSearch - WebFetch permission_mode: acceptEdits - status_rules_prompt: | - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | AI問題の修正完了 | `[CODER:DONE]` | - | 判断できない/情報不足 | `[CODER:BLOCKED]` | - - **重要**: 迷ったら `[BLOCKED]`。勝手に判断しない。 - - ### 出力例 - - **DONE の場合:** - ``` - AI Reviewerの指摘を修正しました。 - - 存在しないAPIを修正: `fetch.json()` → `response.json()` - - [CODER:DONE] - ``` - - **BLOCKED の場合:** - ``` - [CODER:BLOCKED] - - 理由: 指摘された問題の修正方法が不明確です - 必要な情報: 代替APIの仕様 - ``` + rules: + - condition: Architectの指摘を修正完了 + next: review + - condition: 判断できない、情報不足 + next: plan instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: ai_fix - - ## AI Review Feedback (これが最新の指示です - 優先して対応してください) + ## Architect Feedback (これが最新の指示です - 優先して対応してください) {previous_response} - ## Original User Request (ワークフロー開始時の元の要求 - 参考情報) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions - **重要**: AI Reviewerのフィードバックに対応してください。 - 以下に集中してください: - - 間違った仮定の修正 - - もっともらしいが間違っている実装の修正 - - 既存コードベースのパターンとの整合 - - スコープクリープの除去 + **重要**: Architectのフィードバックに対応してください。 + セッションの会話履歴を確認し、Architectの指摘事項を修正してください。 pass_previous_response: true - transitions: - - condition: done - next_step: review - - condition: blocked - next_step: plan - name: security_review - agent: ~/.takt/agents/default/security.md + edit: false + agent: ~/.takt/agents/default/security-reviewer.md + report: + name: 05-security-review.md + format: | + ```markdown + # セキュリティレビュー + + ## 結果: APPROVE / REJECT + + ## 重大度: None / Low / Medium / High / Critical + + ## チェック結果 + | カテゴリ | 結果 | 備考 | + |---------|------|------| + | インジェクション | ✅ | - | + | 認証・認可 | ✅ | - | + | データ保護 | ✅ | - | + | 依存関係 | ✅ | - | + + ## 脆弱性（REJECTの場合） + | # | 重大度 | 種類 | 場所 | 修正案 | + |---|--------|------|------|--------| + | 1 | High | SQLi | `src/db.ts:42` | パラメータ化クエリを使用 | + + ## 警告（ブロッキングではない） + - {セキュリティに関する推奨事項} + ``` + + **認知負荷軽減ルール:** + - 問題なし → チェック表のみ（10行以内） + - 警告あり → + 警告を1-2行（15行以内） + - 脆弱性あり → + 表形式で（30行以内） allowed_tools: - Read - Glob @@ -622,103 +379,20 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - ## 判定基準 - - | 状況 | 判定 | - |------|------| - | 重大な脆弱性 | REJECT | - | 中程度の脆弱性 | REJECT | - | 軽微な問題・警告のみ | APPROVE（警告を付記） | - | セキュリティ問題なし | APPROVE | - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | セキュリティ問題なし | `[SECURITY:APPROVE]` | - | 脆弱性があり修正が必要 | `[SECURITY:REJECT]` | - - ### 出力例 - - **APPROVE の場合:** - ``` - [SECURITY:APPROVE] - - セキュリティ問題は検出されませんでした。 - ``` - - **REJECT の場合:** - ``` - [SECURITY:REJECT] - - 脆弱性: - 1. SQLインジェクション: `src/db.ts:42` - 修正案: パラメータ化クエリを使用 - ``` + rules: + - condition: セキュリティ問題なし + next: supervise + - condition: 脆弱性があり修正が必要 + next: security_fix instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: security_review - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/05-security-review.md - - ## Original User Request (ワークフロー開始時の元の要求) - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions 変更に対してセキュリティレビューを行ってください。以下の脆弱性を確認してください: - インジェクション攻撃（SQL, コマンド, XSS） - 認証・認可の問題 - データ露出リスク - 暗号化の弱点 - **レポート出力:** 上記の `Report File` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - - **レポートフォーマット:** - ```markdown - # セキュリティレビュー - - ## 結果: APPROVE / REJECT - - ## 重大度: None / Low / Medium / High / Critical - - ## チェック結果 - | カテゴリ | 結果 | 備考 | - |---------|------|------| - | インジェクション | ✅ | - | - | 認証・認可 | ✅ | - | - | データ保護 | ✅ | - | - | 依存関係 | ✅ | - | - - ## 脆弱性（REJECTの場合） - | # | 重大度 | 種類 | 場所 | 修正案 | - |---|--------|------|------|--------| - | 1 | High | SQLi | `src/db.ts:42` | パラメータ化クエリを使用 | - - ## 警告（ブロッキングではない） - - {セキュリティに関する推奨事項} - ``` - - **認知負荷軽減ルール:** - - 問題なし → チェック表のみ（10行以内） - - 警告あり → + 警告を1-2行（15行以内） - - 脆弱性あり → + 表形式で（30行以内） - transitions: - - condition: approved - next_step: supervise - - condition: rejected - next_step: security_fix - - name: security_fix + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -730,126 +404,25 @@ steps: - WebSearch - WebFetch permission_mode: acceptEdits - status_rules_prompt: | - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | セキュリティ修正完了 | `[CODER:DONE]` | - | 判断できない/情報不足 | `[CODER:BLOCKED]` | - - **重要**: セキュリティの問題は最優先で対応が必要です。 - - ### 出力例 - - **DONE の場合:** - ``` - セキュリティの指摘を修正しました。 - - SQLインジェクション対策: パラメータ化クエリに変更 - - [CODER:DONE] - ``` - - **BLOCKED の場合:** - ``` - [CODER:BLOCKED] - - 理由: 修正方法が不明確です - 必要な情報: 認証ライブラリの仕様 - ``` + rules: + - condition: セキュリティ修正完了 + next: security_review + - condition: 判断できない、情報不足 + next: plan instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: security_fix - ## Security Review Feedback (これが最新の指示です - 優先して対応してください) {previous_response} - ## Original User Request (ワークフロー開始時の元の要求 - 参考情報) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **重要**: セキュリティレビューで指摘された脆弱性を修正してください。 セキュリティの問題は最優先で対応が必要です。 pass_previous_response: true - transitions: - - condition: done - next_step: security_review - - condition: blocked - next_step: plan - - - name: fix - agent: ~/.takt/agents/default/coder.md - allowed_tools: - - Read - - Glob - - Grep - - Edit - - Write - - Bash - - WebSearch - - WebFetch - permission_mode: acceptEdits - status_rules_prompt: | - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | Architectの指摘を修正完了 | `[CODER:DONE]` | - | 判断できない/情報不足 | `[CODER:BLOCKED]` | - - **重要**: 迷ったら `[BLOCKED]`。勝手に判断しない。 - - ### 出力例 - - **DONE の場合:** - ``` - Architectの指摘を修正しました。 - - ファイル分割: 3ファイルに分割 - - 型定義を追加 - - [CODER:DONE] - ``` - - **BLOCKED の場合:** - ``` - [CODER:BLOCKED] - - 理由: 修正方針が不明確です - 必要な情報: 分割の具体的な方法 - ``` - instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: fix - - ## Architect Feedback (これが最新の指示です - 優先して対応してください) - {previous_response} - - ## Original User Request (ワークフロー開始時の元の要求 - 参考情報) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions - **重要**: Architectのフィードバックに対応してください。 - 「Original User Request」は参考情報であり、最新の指示ではありません。 - セッションの会話履歴を確認し、Architectの指摘事項を修正してください。 - pass_previous_response: true - transitions: - - condition: done - next_step: review - - condition: blocked - next_step: plan - name: supervise + edit: false agent: ~/.takt/agents/default/supervisor.md + report: + - Validation: 06-supervisor-validation.md + - Summary: summary.md allowed_tools: - Read - Glob @@ -858,65 +431,12 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - ## 判定基準 - - | 状況 | 判定 | - |------|------| - | 要求が満たされていない | REJECT | - | テストが失敗する | REJECT | - | ビルドが通らない | REJECT | - | その場しのぎが残っている | REJECT | - | すべて問題なし | APPROVE | - - **原則**: 疑わしきは REJECT。曖昧な承認はしない。 - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 最終承認 | `[SUPERVISOR:APPROVE]` | - | 差し戻し | `[SUPERVISOR:REJECT]` | - - ### 出力例 - - **APPROVE の場合:** - ``` - [SUPERVISOR:APPROVE] - - 検証結果: - - テスト: ✅ 全件パス - - ビルド: ✅ 成功 - - 要求充足: ✅ - ``` - - **REJECT の場合:** - ``` - [SUPERVISOR:REJECT] - - 問題点: - 1. テストが失敗: `npm test` で 2 件失敗 - 2. 元の要求を満たしていない: ログイン機能が未実装 - ``` + rules: + - condition: すべて問題なし + next: COMPLETE + - condition: 要求未達成、テスト失敗、ビルドエラー + next: plan instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: supervise (final verification) - - Report Directory: .takt/reports/{report_dir}/ - - Report Files: - - Validation: .takt/reports/{report_dir}/06-supervisor-validation.md - - Summary: .takt/reports/{report_dir}/summary.md - - ## Original User Request - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions テスト実行、ビルド確認、最終承認を行ってください。 **ワークフロー全体の確認:** @@ -927,10 +447,6 @@ steps: **レポートの確認:** Report Directory内の全レポートを読み、 未対応の改善提案がないか確認してください。 - **レポート出力:** 上記の `Report Files` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - **Validationレポートフォーマット:** ```markdown # 最終検証結果 @@ -984,8 +500,3 @@ steps: npm run build ``` ``` - transitions: - - condition: approved - next_step: COMPLETE - - condition: rejected - next_step: plan diff --git a/resources/global/ja/workflows/expert-cqrs.yaml b/resources/global/ja/workflows/expert-cqrs.yaml index 6c03a76..ba94f6e 100644 --- a/resources/global/ja/workflows/expert-cqrs.yaml +++ b/resources/global/ja/workflows/expert-cqrs.yaml @@ -2,28 +2,36 @@ # CQRS+ES、フロントエンド、セキュリティ、QAの専門家によるレビューワークフロー # # フロー: -# plan -> implement -> architect_review -> cqrs_es_review -> frontend_review -> ai_review -> security_review -> qa_review -> supervise -> COMPLETE -# ↓ ↓ ↓ ↓ ↓ ↓ ↓ -# fix_architect fix_cqrs_es fix_frontend ai_fix fix_security fix_qa fix_supervisor +# plan -> implement -> ai_review -> cqrs_es_review -> frontend_review -> security_review -> qa_review -> supervise -> COMPLETE +# ↓ ↓ ↓ ↓ ↓ ↓ +# ai_fix fix_cqrs_es fix_frontend fix_security fix_qa fix_supervisor # # 修正時の戻り先はCoderが判断: # - fix_security: MINOR→security_review, MAJOR→cqrs_es_review # - fix_qa: MINOR→qa_review, SECURITY→security_review, MAJOR→cqrs_es_review # -# テンプレート変数: -# {iteration} - ワークフロー全体のターン数（全エージェントで実行されたステップの合計） -# {max_iterations} - ワークフローの最大イテレーション数 -# {step_iteration} - ステップごとのイテレーション数（このステップが何回実行されたか） -# {task} - 元のユーザー要求 -# {previous_response} - 前のステップの出力 -# {git_diff} - 現在のコミットされていない変更（git diff） -# {user_inputs} - ワークフロー中に蓄積されたユーザー入力 -# {report_dir} - レポートディレクトリ名（例: "20250126-143052-task-summary"） +# ボイラープレートセクション（Workflow Context, User Request, Previous Response, +# Additional User Inputs, Instructions heading）はbuildInstruction()が自動挿入。 +# instruction_templateにはステップ固有の内容のみ記述。 +# +# テンプレート変数（instruction_template内で使用可能）: +# {iteration} - ワークフロー全体のターン数（全エージェントで実行されたステップの合計） +# {max_iterations} - ワークフローの最大イテレーション数 +# {step_iteration} - ステップごとのイテレーション数（このステップが何回実行されたか） +# {previous_response} - 前のステップの出力（pass_previous_response: true の場合のみ） +# {report_dir} - レポートディレクトリ名（例: "20250126-143052-task-summary"） +# +# ステップレベルフィールド: +# report: - ステップのレポートファイル（Workflow ContextにReport File/Filesとして自動挿入） +# 単一: report: 00-plan.md +# 複数: report: +# - Scope: 01-coder-scope.md +# - Decisions: 02-coder-decisions.md name: expert-cqrs description: CQRS+ES・フロントエンド・セキュリティ・QA専門家レビュー -max_iterations: 20 +max_iterations: 30 initial_step: plan @@ -32,7 +40,31 @@ steps: # Phase 0: Planning # =========================================== - name: plan + edit: false agent: ~/.takt/agents/default/planner.md + report: + name: 00-plan.md + format: | + ```markdown + # タスク計画 + + ## 元の要求 + {ユーザーの要求をそのまま記載} + + ## 分析結果 + + ### 目的 + {達成すべきこと} + + ### スコープ + {影響範囲} + + ### 実装アプローチ + {どう進めるか} + + ## 確認事項（あれば） + - {不明点や確認が必要な点} + ``` allowed_tools: - Read - Glob @@ -41,30 +73,10 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 分析完了 | `[PLANNER:DONE]` | - | 要件不明確 | `[PLANNER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: plan (タスク分析) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/00-plan.md - - ## User Request - {task} - ## Previous Response (implementからの差し戻し時) {previous_response} - ## Instructions タスクを分析し、実装方針を立ててください。 **注意:** Previous Responseがある場合は差し戻しのため、 @@ -74,47 +86,22 @@ steps: 1. タスクの要件を理解する 2. 影響範囲を特定する 3. 実装アプローチを決める - - **レポート出力:** 上記の `Report File` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - - **レポートフォーマット:** - ```markdown - # タスク計画 - - ## 元の要求 - {ユーザーの要求をそのまま記載} - - ## 分析結果 - - ### 目的 - {達成すべきこと} - - ### スコープ - {影響範囲} - - ### 実装アプローチ - {どう進めるか} - - ## 確認事項（あれば） - - {不明点や確認が必要な点} - ``` - - 完了したら [PLANNER:DONE] を出力。 - 要件が不明確な場合は [PLANNER:BLOCKED] を出力。 pass_previous_response: true - transitions: - - condition: done - next_step: implement - - condition: blocked - next_step: ABORT + rules: + - condition: タスク分析と計画が完了した + next: implement + - condition: 要件が不明確で計画を立てられない + next: ABORT # =========================================== # Phase 1: Implementation # =========================================== - name: implement + edit: true agent: ~/.takt/agents/default/coder.md + report: + - Scope: 01-coder-scope.md + - Decisions: 02-coder-decisions.md allowed_tools: - Read - Glob @@ -124,39 +111,10 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 実装完了 | `[CODER:DONE]` | - | 進行不可 | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: implement - - Report Directory: .takt/reports/{report_dir}/ - - Report Files: - - Scope: .takt/reports/{report_dir}/01-coder-scope.md - - Decisions: .takt/reports/{report_dir}/02-coder-decisions.md - - ## User Request - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions planステップで立てた計画に従って実装してください。 計画レポート（00-plan.md）を参照し、実装を進めてください。 - **レポート出力:** 上記の `Report Files` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - **Scopeレポートフォーマット（実装開始時に作成）:** ```markdown # 変更スコープ宣言 @@ -186,20 +144,46 @@ steps: - **検討した選択肢**: {選択肢リスト} - **理由**: {選んだ理由} ``` - - 完了時は [CODER:DONE] を含めてください。 - 進行できない場合は [CODER:BLOCKED] を出力し、planに戻ります。 - transitions: - - condition: done - next_step: architect_review - - condition: blocked - next_step: plan + rules: + - condition: 実装が完了した + next: ai_review + - condition: 実装を進行できない + next: plan # =========================================== - # Phase 2: Architecture Review + # Phase 2: AI Review # =========================================== - - name: architect_review - agent: ~/.takt/agents/default/architect.md + - name: ai_review + edit: false + agent: ~/.takt/agents/default/ai-antipattern-reviewer.md + report: + name: 03-ai-review.md + format: | + ```markdown + # AI生成コードレビュー + + ## 結果: APPROVE / REJECT + + ## サマリー + {1文で結果を要約} + + ## 検証した項目 + | 観点 | 結果 | 備考 | + |------|------|------| + | 仮定の妥当性 | ✅ | - | + | API/ライブラリの実在 | ✅ | - | + | コンテキスト適合 | ✅ | - | + | スコープ | ✅ | - | + + ## 問題点（REJECTの場合） + | # | カテゴリ | 場所 | 問題 | + |---|---------|------|------| + | 1 | 幻覚API | `src/file.ts:23` | 存在しないメソッド | + ``` + + **認知負荷軽減ルール:** + - 問題なし → サマリー1文 + チェック表のみ（10行以内） + - 問題あり → + 問題を表形式で（25行以内） allowed_tools: - Read - Glob @@ -207,97 +191,20 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - - - ## 判定基準 - - | 状況 | 判定 | - |------|------| - | 構造に問題がある | REJECT | - | 設計原則違反がある | REJECT | - | 呼び出しチェーンの配線漏れ | REJECT | - | テストが不十分 | REJECT | - | 改善すべき点がある（軽微） | IMPROVE | - | 問題なし | APPROVE | - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 問題なし | `[ARCHITECT:APPROVE]` | - | 軽微な改善必要 | `[ARCHITECT:IMPROVE]` | - | 構造的な修正必要 | `[ARCHITECT:REJECT]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: architect_review (アーキテクチャレビュー) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/03-architect-review.md + AI特有の問題についてコードをレビューしてください: + - 仮定の検証 + - もっともらしいが間違っているパターン + - 既存コードベースとの適合性 + - スコープクリープの検出 + rules: + - condition: AI特有の問題が見つからない + next: cqrs_es_review + - condition: AI特有の問題が検出された + next: ai_fix - ## Original User Request (ワークフロー開始時の元の要求) - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions - **アーキテクチャと設計**のレビューに集中してください。 - - **レビュー観点:** - - 構造・設計の妥当性 - - コード品質 - - 変更スコープの適切性 - - テストカバレッジ - - デッドコード - - 呼び出しチェーン検証 - - **レポート出力:** 上記の `Report File` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - - **レポートフォーマット:** - ```markdown - # アーキテクチャレビュー - - ## 結果: APPROVE / IMPROVE / REJECT - - ## サマリー - {1-2文で結果を要約} - - ## 確認した観点 - - [x] 構造・設計 - - [x] コード品質 - - [x] 変更スコープ - - [x] テストカバレッジ - - [x] デッドコード - - [x] 呼び出しチェーン検証 - - ## 問題点（REJECTの場合） - | # | 場所 | 問題 | 修正案 | - |---|------|------|--------| - | 1 | `src/file.ts:42` | 問題の説明 | 修正方法 | - - ## 改善提案（任意・ブロッキングではない） - - {将来的な改善提案} - ``` - - **認知負荷軽減ルール:** - - APPROVE + 問題なし → サマリーのみ（5行以内） - - APPROVE + 軽微な提案 → サマリー + 改善提案（15行以内） - - REJECT → 問題点を表形式で（30行以内） - transitions: - - condition: approved - next_step: cqrs_es_review - - condition: improve - next_step: fix_architect - - condition: rejected - next_step: fix_architect - - - name: fix_architect + - name: ai_fix + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -308,50 +215,54 @@ steps: - Bash - WebSearch - WebFetch - permission_mode: acceptEdits - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 修正完了 | `[CODER:DONE]` | - | 進行不可 | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: fix_architect - - ## Architect Feedback (これが最新の指示です - 優先して対応してください) + ## AI Review Feedback (これが最新の指示です - 優先して対応してください) {previous_response} - ## Original User Request (ワークフロー開始時の元の要求 - 参考情報) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions - **重要**: Architectのフィードバックに対応してください。 - 「Original User Request」は参考情報であり、最新の指示ではありません。 - セッションの会話履歴を確認し、Architectの指摘事項を修正してください。 - - 完了時は [CODER:DONE] を含めてください。 - 進行できない場合は [CODER:BLOCKED] を含めてください。 + **重要**: AI Reviewerのフィードバックに対応してください。 + 以下に集中してください: + - 間違った仮定の修正 + - もっともらしいが間違っている実装の修正 + - 既存コードベースのパターンとの整合 + - スコープクリープの除去 pass_previous_response: true - transitions: - - condition: done - next_step: architect_review - - condition: blocked - next_step: plan + rules: + - condition: AI Reviewerの指摘に対する修正が完了した + next: ai_review + - condition: 修正を進行できない + next: plan # =========================================== # Phase 3: CQRS+ES Review # =========================================== - name: cqrs_es_review + edit: false agent: ~/.takt/agents/expert-cqrs/cqrs-es-reviewer.md + report: + name: 04-cqrs-es-review.md + format: | + ```markdown + # CQRS+ESレビュー + + ## 結果: APPROVE / REJECT + + ## サマリー + {1-2文で結果を要約} + + ## 確認した観点 + | 観点 | 結果 | 備考 | + |------|------|------| + | Aggregate設計 | ✅ | - | + | イベント設計 | ✅ | - | + | Command/Query分離 | ✅ | - | + | プロジェクション | ✅ | - | + | 結果整合性 | ✅ | - | + + ## 問題点（REJECTの場合） + | # | 場所 | 問題 | 修正案 | + |---|------|------|--------| + | 1 | `src/file.ts:42` | 問題の説明 | 修正方法 | + ``` allowed_tools: - Read - Glob @@ -359,32 +270,7 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 設計に問題なし | `[CQRS-ES:APPROVE]` | - | 設計に問題あり | `[CQRS-ES:REJECT]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: cqrs_es_review (CQRS+ES専門レビュー) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/04-cqrs-es-review.md - - ## Original User Request - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions CQRS（コマンドクエリ責務分離）とEvent Sourcing（イベントソーシング）の観点から 上記の変更をレビューしてください。 @@ -397,45 +283,14 @@ steps: **注意**: このプロジェクトがCQRS+ESパターンを使用していない場合は、 一般的なドメイン設計の観点からレビューしてください。 - - **レポート出力:** 上記の `Report File` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - - **レポートフォーマット:** - ```markdown - # CQRS+ESレビュー - - ## 結果: APPROVE / REJECT - - ## サマリー - {1-2文で結果を要約} - - ## 確認した観点 - | 観点 | 結果 | 備考 | - |------|------|------| - | Aggregate設計 | ✅ | - | - | イベント設計 | ✅ | - | - | Command/Query分離 | ✅ | - | - | プロジェクション | ✅ | - | - | 結果整合性 | ✅ | - | - - ## 問題点（REJECTの場合） - | # | 場所 | 問題 | 修正案 | - |---|------|------|--------| - | 1 | `src/file.ts:42` | 問題の説明 | 修正方法 | - ``` - - Include: - - [CQRS-ES:APPROVE] if CQRS+ES design is sound - - [CQRS-ES:REJECT] if design issues found (list specific issues) - transitions: - - condition: approved - next_step: frontend_review - - condition: rejected - next_step: fix_cqrs_es + rules: + - condition: CQRS+ES設計に問題がない + next: frontend_review + - condition: CQRS+ES設計に問題がある + next: fix_cqrs_es - name: fix_cqrs_es + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -446,31 +301,10 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 修正完了 | `[CODER:DONE]` | - | 進行不可 | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: fix_cqrs_es - ## CQRS+ES Review Feedback (これが最新の指示です - 優先して対応してください) {previous_response} - ## Original User Request (ワークフロー開始時の元の要求 - 参考情報) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **重要**: CQRS+ES専門家からの指摘を修正してください。 指摘されたポイント: @@ -479,21 +313,44 @@ steps: - Command/Query分離 - プロジェクション - 結果整合性 - - 完了時は [CODER:DONE] を含めてください。 - 進行できない場合は [CODER:BLOCKED] を含めてください。 pass_previous_response: true - transitions: - - condition: done - next_step: cqrs_es_review - - condition: blocked - next_step: plan + rules: + - condition: CQRS+ES専門家の指摘に対する修正が完了した + next: cqrs_es_review + - condition: 修正を進行できない + next: plan # =========================================== - # Phase 3: Frontend Review + # Phase 4: Frontend Review # =========================================== - name: frontend_review + edit: false agent: ~/.takt/agents/expert/frontend-reviewer.md + report: + name: 05-frontend-review.md + format: | + ```markdown + # フロントエンドレビュー + + ## 結果: APPROVE / REJECT + + ## サマリー + {1-2文で結果を要約} + + ## 確認した観点 + | 観点 | 結果 | 備考 | + |------|------|------| + | コンポーネント設計 | ✅ | - | + | 状態管理 | ✅ | - | + | パフォーマンス | ✅ | - | + | アクセシビリティ | ✅ | - | + | 型安全性 | ✅ | - | + + ## 問題点（REJECTの場合） + | # | 場所 | 問題 | 修正案 | + |---|------|------|--------| + | 1 | `src/file.tsx:42` | 問題の説明 | 修正方法 | + ``` allowed_tools: - Read - Glob @@ -501,32 +358,7 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | フロントエンド設計に問題なし | `[FRONTEND:APPROVE]` | - | フロントエンド設計に問題あり | `[FRONTEND:REJECT]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: frontend_review (フロントエンド専門レビュー) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/05-frontend-review.md - - ## Original User Request - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions フロントエンド開発の観点から上記の変更をレビューしてください。 **レビュー観点:** @@ -538,46 +370,15 @@ steps: - TypeScript型安全性 **注意**: このプロジェクトがフロントエンドを含まない場合は、 - [FRONTEND:APPROVE] として次に進んでください。 - - **レポート出力:** 上記の `Report File` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - - **レポートフォーマット:** - ```markdown - # フロントエンドレビュー - - ## 結果: APPROVE / REJECT - - ## サマリー - {1-2文で結果を要約} - - ## 確認した観点 - | 観点 | 結果 | 備考 | - |------|------|------| - | コンポーネント設計 | ✅ | - | - | 状態管理 | ✅ | - | - | パフォーマンス | ✅ | - | - | アクセシビリティ | ✅ | - | - | 型安全性 | ✅ | - | - - ## 問題点（REJECTの場合） - | # | 場所 | 問題 | 修正案 | - |---|------|------|--------| - | 1 | `src/file.tsx:42` | 問題の説明 | 修正方法 | - ``` - - Include: - - [FRONTEND:APPROVE] if frontend design is sound - - [FRONTEND:REJECT] if design issues found (list specific issues) - transitions: - - condition: approved - next_step: ai_review - - condition: rejected - next_step: fix_frontend + 問題なしとして次に進んでください。 + rules: + - condition: フロントエンド設計に問題がない + next: security_review + - condition: フロントエンド設計に問題がある + next: fix_frontend - name: fix_frontend + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -588,31 +389,10 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 修正完了 | `[CODER:DONE]` | - | 進行不可 | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: fix_frontend - ## Frontend Review Feedback (これが最新の指示です - 優先して対応してください) {previous_response} - ## Original User Request (ワークフロー開始時の元の要求 - 参考情報) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **重要**: フロントエンド専門家からの指摘を修正してください。 指摘されたポイント: @@ -621,157 +401,45 @@ steps: - パフォーマンス - アクセシビリティ - 型安全性 - - 完了時は [CODER:DONE] を含めてください。 - 進行できない場合は [CODER:BLOCKED] を含めてください。 pass_previous_response: true - transitions: - - condition: done - next_step: frontend_review - - condition: blocked - next_step: plan - - # =========================================== - # Phase 4: AI Review - # =========================================== - - name: ai_review - agent: ~/.takt/agents/default/ai-antipattern-reviewer.md - allowed_tools: - - Read - - Glob - - Grep - - Write - - WebSearch - - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | AI特有の問題なし | `[AI_REVIEW:APPROVE]` | - | 問題あり | `[AI_REVIEW:REJECT]` | - instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: ai_review (AI生成コードレビュー) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/06-ai-review.md - - ## Original User Request (ワークフロー開始時の元の要求) - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions - AI特有の問題についてコードをレビューしてください: - - 仮定の検証 - - もっともらしいが間違っているパターン - - 既存コードベースとの適合性 - - スコープクリープの検出 - - **レポート出力:** 上記の `Report File` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - - **レポートフォーマット:** - ```markdown - # AI生成コードレビュー - - ## 結果: APPROVE / REJECT - - ## サマリー - {1文で結果を要約} - - ## 検証した項目 - | 観点 | 結果 | 備考 | - |------|------|------| - | 仮定の妥当性 | ✅ | - | - | API/ライブラリの実在 | ✅ | - | - | コンテキスト適合 | ✅ | - | - | スコープ | ✅ | - | - - ## 問題点（REJECTの場合） - | # | カテゴリ | 場所 | 問題 | - |---|---------|------|------| - | 1 | 幻覚API | `src/file.ts:23` | 存在しないメソッド | - ``` - - **認知負荷軽減ルール:** - - 問題なし → サマリー1文 + チェック表のみ（10行以内） - - 問題あり → + 問題を表形式で（25行以内） - - 以下を含めてください: - - [AI_REVIEW:APPROVE] AI特有の問題が見つからない場合 - - [AI_REVIEW:REJECT] 問題が検出された場合（具体的な問題をリスト） - transitions: - - condition: approved - next_step: security_review - - condition: rejected - next_step: ai_fix - - - name: ai_fix - agent: ~/.takt/agents/default/coder.md - allowed_tools: - - Read - - Glob - - Grep - - Edit - - Write - - Bash - - WebSearch - - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 修正完了 | `[CODER:DONE]` | - | 進行不可 | `[CODER:BLOCKED]` | - instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: ai_fix - - ## AI Review Feedback (これが最新の指示です - 優先して対応してください) - {previous_response} - - ## Original User Request (ワークフロー開始時の元の要求 - 参考情報) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions - **重要**: AI Reviewerのフィードバックに対応してください。 - 以下に集中してください: - - 間違った仮定の修正 - - もっともらしいが間違っている実装の修正 - - 既存コードベースのパターンとの整合 - - スコープクリープの除去 - - 完了時は [CODER:DONE] を含めてください。 - 進行できない場合は [CODER:BLOCKED] を含めてください。 - pass_previous_response: true - transitions: - - condition: done - next_step: ai_review - - condition: blocked - next_step: plan + rules: + - condition: フロントエンドの指摘に対する修正が完了した + next: frontend_review + - condition: 修正を進行できない + next: plan # =========================================== # Phase 5: Security Review # =========================================== - name: security_review + edit: false agent: ~/.takt/agents/expert/security-reviewer.md + report: + name: 06-security-review.md + format: | + ```markdown + # セキュリティレビュー + + ## 結果: APPROVE / REJECT + + ## 重大度: None / Low / Medium / High / Critical + + ## チェック結果 + | カテゴリ | 結果 | 備考 | + |---------|------|------| + | インジェクション | ✅ | - | + | 認証・認可 | ✅ | - | + | データ保護 | ✅ | - | + | 依存関係 | ✅ | - | + + ## 脆弱性（REJECTの場合） + | # | 重大度 | 種類 | 場所 | 修正案 | + |---|--------|------|------|--------| + | 1 | High | SQLi | `src/db.ts:42` | パラメータ化クエリを使用 | + + ## 警告（ブロッキングではない） + - {セキュリティに関する推奨事項} + ``` allowed_tools: - Read - Glob @@ -779,32 +447,7 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | セキュリティ問題なし | `[SECURITY:APPROVE]` | - | 脆弱性あり | `[SECURITY:REJECT]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: security_review (セキュリティ専門レビュー) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/07-security-review.md - - ## Original User Request - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions セキュリティの観点から上記の変更をレビューしてください。 **レビュー観点:** @@ -813,46 +456,14 @@ steps: - 機密情報の取り扱い - 暗号化の適切性 - OWASP Top 10 - - **レポート出力:** 上記の `Report File` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - - **レポートフォーマット:** - ```markdown - # セキュリティレビュー - - ## 結果: APPROVE / REJECT - - ## 重大度: None / Low / Medium / High / Critical - - ## チェック結果 - | カテゴリ | 結果 | 備考 | - |---------|------|------| - | インジェクション | ✅ | - | - | 認証・認可 | ✅ | - | - | データ保護 | ✅ | - | - | 依存関係 | ✅ | - | - - ## 脆弱性（REJECTの場合） - | # | 重大度 | 種類 | 場所 | 修正案 | - |---|--------|------|------|--------| - | 1 | High | SQLi | `src/db.ts:42` | パラメータ化クエリを使用 | - - ## 警告（ブロッキングではない） - - {セキュリティに関する推奨事項} - ``` - - Include: - - [SECURITY:APPROVE] if no security issues found - - [SECURITY:REJECT] if vulnerabilities found (list specific issues with severity) - transitions: - - condition: approved - next_step: qa_review - - condition: rejected - next_step: fix_security + rules: + - condition: セキュリティ上の問題がない + next: qa_review + - condition: セキュリティ上の脆弱性が検出された + next: fix_security - name: fix_security + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -863,32 +474,10 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 軽微な修正完了 | `[CODER:DONE]` | - | 大きな修正（CQRS+ESからやり直し） | `[CODER:REJECT]` | - | 進行不可 | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: fix_security - ## Security Review Feedback (これが最新の指示です - 優先して対応してください) {previous_response} - ## Original User Request (ワークフロー開始時の元の要求 - 参考情報) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **重要**: セキュリティ専門家からの指摘を修正してください。 セキュリティ問題は最優先で対応してください。 @@ -899,28 +488,52 @@ steps: - 暗号化の問題 ## 修正完了時の判断 - 修正が完了したら、**変更の影響範囲**を判断して適切なタグを出力してください： + 修正が完了したら、**変更の影響範囲**を判断してください： - - `[CODER:DONE]` - 軽微な修正（セキュリティレビューのみ再実施） + - 軽微な修正（セキュリティレビューのみ再実施） - 例: バリデーション追加、エスケープ処理追加、設定変更 - - `[CODER:REJECT]` - 大きな修正（CQRS+ESレビューからやり直し） + - 大きな修正（CQRS+ESレビューからやり直し） - 例: データフロー変更、API設計変更、認証方式変更、ドメインモデル変更 - - 進行できない場合は [CODER:BLOCKED] を含めてください。 pass_previous_response: true - transitions: - - condition: done - next_step: security_review - - condition: rejected - next_step: cqrs_es_review - - condition: blocked - next_step: plan + rules: + - condition: 軽微なセキュリティ修正が完了した + next: security_review + - condition: 大きな修正を行い、CQRS+ESレビューからやり直す必要がある + next: cqrs_es_review + - condition: 修正を進行できない + next: plan # =========================================== # Phase 6: QA Review # =========================================== - name: qa_review + edit: false agent: ~/.takt/agents/expert/qa-reviewer.md + report: + name: 07-qa-review.md + format: | + ```markdown + # QAレビュー + + ## 結果: APPROVE / REJECT + + ## サマリー + {1-2文で結果を要約} + + ## 確認した観点 + | 観点 | 結果 | 備考 | + |------|------|------| + | テストカバレッジ | ✅ | - | + | テスト品質 | ✅ | - | + | エラーハンドリング | ✅ | - | + | ドキュメント | ✅ | - | + | 保守性 | ✅ | - | + + ## 問題点（REJECTの場合） + | # | カテゴリ | 問題 | 修正案 | + |---|---------|------|--------| + | 1 | テスト | 問題の説明 | 修正方法 | + ``` allowed_tools: - Read - Glob @@ -928,32 +541,7 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 品質基準を満たす | `[QA:APPROVE]` | - | 品質問題あり | `[QA:REJECT]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: qa_review (QA専門レビュー) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/08-qa-review.md - - ## Original User Request - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions 品質保証の観点から上記の変更をレビューしてください。 **レビュー観点:** @@ -963,45 +551,14 @@ steps: - エラーハンドリング - ログとモニタリング - 保守性 - - **レポート出力:** 上記の `Report File` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - - **レポートフォーマット:** - ```markdown - # QAレビュー - - ## 結果: APPROVE / REJECT - - ## サマリー - {1-2文で結果を要約} - - ## 確認した観点 - | 観点 | 結果 | 備考 | - |------|------|------| - | テストカバレッジ | ✅ | - | - | テスト品質 | ✅ | - | - | エラーハンドリング | ✅ | - | - | ドキュメント | ✅ | - | - | 保守性 | ✅ | - | - - ## 問題点（REJECTの場合） - | # | カテゴリ | 問題 | 修正案 | - |---|---------|------|--------| - | 1 | テスト | 問題の説明 | 修正方法 | - ``` - - Include: - - [QA:APPROVE] if quality standards are met - - [QA:REJECT] if quality issues found (list specific issues) - transitions: - - condition: approved - next_step: supervise - - condition: rejected - next_step: fix_qa + rules: + - condition: 品質基準を満たしている + next: supervise + - condition: 品質に問題がある + next: fix_qa - name: fix_qa + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -1012,33 +569,10 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 軽微な修正完了 | `[CODER:DONE]` | - | セキュリティ影響あり | `[CODER:IMPROVE]` | - | 大きな修正（CQRS+ESからやり直し） | `[CODER:REJECT]` | - | 進行不可 | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: fix_qa - ## QA Review Feedback (これが最新の指示です - 優先して対応してください) {previous_response} - ## Original User Request (ワークフロー開始時の元の要求 - 参考情報) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **重要**: QA専門家からの指摘を修正してください。 指摘されたポイント: @@ -1049,32 +583,34 @@ steps: - コード品質 ## 修正完了時の判断 - 修正が完了したら、**変更の影響範囲**を判断して適切なタグを出力してください： + 修正が完了したら、**変更の影響範囲**を判断してください： - - `[CODER:DONE]` - 軽微な修正（QAレビューのみ再実施） + - 軽微な修正（QAレビューのみ再実施） - 例: テスト追加、ドキュメント追加、ログ追加、コメント追加 - - `[CODER:IMPROVE]` - セキュリティに影響する修正（セキュリティレビューからやり直し） + - セキュリティに影響する修正（セキュリティレビューからやり直し） - 例: エラーハンドリング変更（エラーメッセージの内容変更）、入力検証の変更 - - `[CODER:REJECT]` - 大きな修正（CQRS+ESレビューからやり直し） + - 大きな修正（CQRS+ESレビューからやり直し） - 例: ビジネスロジック変更、データモデル変更、API変更 - - 進行できない場合は [CODER:BLOCKED] を含めてください。 pass_previous_response: true - transitions: - - condition: done - next_step: qa_review - - condition: improve - next_step: security_review - - condition: rejected - next_step: cqrs_es_review - - condition: blocked - next_step: plan + rules: + - condition: 軽微なQA修正が完了した + next: qa_review + - condition: セキュリティに影響する修正を行った + next: security_review + - condition: 大きな修正を行い、CQRS+ESレビューからやり直す必要がある + next: cqrs_es_review + - condition: 修正を進行できない + next: plan # =========================================== # Phase 7: Supervision # =========================================== - name: supervise + edit: false agent: ~/.takt/agents/expert/supervisor.md + report: + - Validation: 08-supervisor-validation.md + - Summary: summary.md allowed_tools: - Read - Glob @@ -1082,43 +618,15 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 最終承認 | `[SUPERVISOR:APPROVE]` | - | 問題あり | `[SUPERVISOR:REJECT]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: supervise (最終確認) - - Report Directory: .takt/reports/{report_dir}/ - - Report Files: - - Validation: .takt/reports/{report_dir}/09-supervisor-validation.md - - Summary: .takt/reports/{report_dir}/summary.md - - ## Original User Request - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - ## Previous Reviews Summary このステップに到達したということは、以下のレビューがすべてAPPROVEされています： - - Architecture Review: APPROVED + - AI Review: APPROVED - CQRS+ES Review: APPROVED - Frontend Review: APPROVED - - AI Review: APPROVED - Security Review: APPROVED - QA Review: APPROVED - ## Instructions テスト実行、ビルド確認、最終承認を行ってください。 **ワークフロー全体の確認:** @@ -1129,10 +637,6 @@ steps: **レポートの確認:** Report Directory内の全レポートを読み、 未対応の改善提案がないか確認してください。 - **レポート出力:** 上記の `Report Files` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - **Validationレポートフォーマット:** ```markdown # 最終検証結果 @@ -1175,10 +679,9 @@ steps: ## レビュー結果 | レビュー | 結果 | |---------|------| - | Architecture | ✅ APPROVE | + | AI Review | ✅ APPROVE | | CQRS+ES | ✅ APPROVE | | Frontend | ✅ APPROVE | - | AI Review | ✅ APPROVE | | Security | ✅ APPROVE | | QA | ✅ APPROVE | | Supervisor | ✅ APPROVE | @@ -1189,17 +692,14 @@ steps: npm run build ``` ``` - - 出力: - - [SUPERVISOR:APPROVE] すべて完了、マージ可能 - - [SUPERVISOR:REJECT] 問題あり（具体的な問題を記載） - transitions: - - condition: approved - next_step: COMPLETE - - condition: rejected - next_step: fix_supervisor + rules: + - condition: すべての検証が完了し、マージ可能な状態である + next: COMPLETE + - condition: 問題が検出された + next: fix_supervisor - name: fix_supervisor + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -1210,41 +710,17 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 修正完了 | `[CODER:DONE]` | - | 進行不可 | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: fix_supervisor - ## Supervisor Feedback (これが最新の指示です - 優先して対応してください) {previous_response} - ## Original User Request (ワークフロー開始時の元の要求 - 参考情報) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **重要**: 監督者からの指摘を修正してください。 監督者は全体を俯瞰した視点から問題を指摘しています。 優先度の高い項目から順に対応してください。 - - 完了時は [CODER:DONE] を含めてください。 - 進行できない場合は [CODER:BLOCKED] を含めてください。 pass_previous_response: true - transitions: - - condition: done - next_step: supervise - - condition: blocked - next_step: plan + rules: + - condition: 監督者の指摘に対する修正が完了した + next: supervise + - condition: 修正を進行できない + next: plan diff --git a/resources/global/ja/workflows/expert.yaml b/resources/global/ja/workflows/expert.yaml index 769ad68..2084468 100644 --- a/resources/global/ja/workflows/expert.yaml +++ b/resources/global/ja/workflows/expert.yaml @@ -2,9 +2,9 @@ # アーキテクチャ、フロントエンド、セキュリティ、QAの専門家によるレビューワークフロー # # フロー: -# plan -> implement -> architect_review -> frontend_review -> ai_review -> security_review -> qa_review -> supervise -> COMPLETE -# ↓ ↓ ↓ ↓ ↓ ↓ -# fix_architect fix_frontend ai_fix fix_security fix_qa fix_supervisor +# plan -> implement -> ai_review -> architect_review -> frontend_review -> security_review -> qa_review -> supervise -> COMPLETE +# ↓ ↓ ↓ ↓ ↓ ↓ +# ai_fix fix_architect fix_frontend fix_security fix_qa fix_supervisor # # 修正時の戻り先はCoderが判断: # - fix_security: MINOR→security_review, MAJOR→architect_review @@ -16,14 +16,13 @@ # {step_iteration} - ステップごとのイテレーション数（このステップが何回実行されたか） # {task} - 元のユーザー要求 # {previous_response} - 前のステップの出力 -# {git_diff} - 現在のコミットされていない変更（git diff） # {user_inputs} - ワークフロー中に蓄積されたユーザー入力 # {report_dir} - レポートディレクトリ名（例: "20250126-143052-task-summary"） name: expert description: アーキテクチャ・フロントエンド・セキュリティ・QA専門家レビュー -max_iterations: 20 +max_iterations: 30 initial_step: plan @@ -32,7 +31,31 @@ steps: # Phase 0: Planning # =========================================== - name: plan + edit: false agent: ~/.takt/agents/default/planner.md + report: + name: 00-plan.md + format: | + ```markdown + # タスク計画 + + ## 元の要求 + {ユーザーの要求をそのまま記載} + + ## 分析結果 + + ### 目的 + {達成すべきこと} + + ### スコープ + {影響範囲} + + ### 実装アプローチ + {どう進めるか} + + ## 確認事項（あれば） + - {不明点や確認が必要な点} + ``` allowed_tools: - Read - Glob @@ -41,30 +64,10 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 分析完了 | `[PLANNER:DONE]` | - | 要件不明確 | `[PLANNER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: plan (タスク分析) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/00-plan.md - - ## User Request - {task} - ## Previous Response (implementからの差し戻し時) {previous_response} - ## Instructions タスクを分析し、実装方針を立ててください。 **注意:** Previous Responseがある場合は差し戻しのため、 @@ -74,47 +77,22 @@ steps: 1. タスクの要件を理解する 2. 影響範囲を特定する 3. 実装アプローチを決める - - **レポート出力:** 上記の `Report File` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - - **レポートフォーマット:** - ```markdown - # タスク計画 - - ## 元の要求 - {ユーザーの要求をそのまま記載} - - ## 分析結果 - - ### 目的 - {達成すべきこと} - - ### スコープ - {影響範囲} - - ### 実装アプローチ - {どう進めるか} - - ## 確認事項（あれば） - - {不明点や確認が必要な点} - ``` - - 完了したら [PLANNER:DONE] を出力。 - 要件が不明確な場合は [PLANNER:BLOCKED] を出力。 pass_previous_response: true - transitions: - - condition: done - next_step: implement - - condition: blocked - next_step: ABORT + rules: + - condition: タスク分析と計画が完了した + next: implement + - condition: 要件が不明確で計画を立てられない + next: ABORT # =========================================== # Phase 1: Implementation # =========================================== - name: implement + edit: true agent: ~/.takt/agents/default/coder.md + report: + - Scope: 01-coder-scope.md + - Decisions: 02-coder-decisions.md allowed_tools: - Read - Glob @@ -124,39 +102,10 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 実装完了 | `[CODER:DONE]` | - | 進行不可 | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: implement - - Report Directory: .takt/reports/{report_dir}/ - - Report Files: - - Scope: .takt/reports/{report_dir}/01-coder-scope.md - - Decisions: .takt/reports/{report_dir}/02-coder-decisions.md - - ## User Request - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions planステップで立てた計画に従って実装してください。 計画レポート（00-plan.md）を参照し、実装を進めてください。 - **レポート出力:** 上記の `Report Files` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - **Scopeレポートフォーマット（実装開始時に作成）:** ```markdown # 変更スコープ宣言 @@ -186,20 +135,46 @@ steps: - **検討した選択肢**: {選択肢リスト} - **理由**: {選んだ理由} ``` - - 完了時は [CODER:DONE] を含めてください。 - 進行できない場合は [CODER:BLOCKED] を出力し、planに戻ります。 - transitions: - - condition: done - next_step: architect_review - - condition: blocked - next_step: plan + rules: + - condition: 実装が完了した + next: ai_review + - condition: 実装を進行できない + next: plan # =========================================== - # Phase 2: Architecture Review + # Phase 2: AI Review # =========================================== - - name: architect_review - agent: ~/.takt/agents/default/architect.md + - name: ai_review + edit: false + agent: ~/.takt/agents/default/ai-antipattern-reviewer.md + report: + name: 03-ai-review.md + format: | + ```markdown + # AI生成コードレビュー + + ## 結果: APPROVE / REJECT + + ## サマリー + {1文で結果を要約} + + ## 検証した項目 + | 観点 | 結果 | 備考 | + |------|------|------| + | 仮定の妥当性 | ✅ | - | + | API/ライブラリの実在 | ✅ | - | + | コンテキスト適合 | ✅ | - | + | スコープ | ✅ | - | + + ## 問題点（REJECTの場合） + | # | カテゴリ | 場所 | 問題 | + |---|---------|------|------| + | 1 | 幻覚API | `src/file.ts:23` | 存在しないメソッド | + ``` + + **認知負荷軽減ルール:** + - 問題なし → サマリー1文 + チェック表のみ（10行以内） + - 問題あり → + 問題を表形式で（25行以内） allowed_tools: - Read - Glob @@ -207,44 +182,93 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - - - ## 判定基準 - - | 状況 | 判定 | - |------|------| - | 構造に問題がある | REJECT | - | 設計原則違反がある | REJECT | - | 呼び出しチェーンの配線漏れ | REJECT | - | テストが不十分 | REJECT | - | 改善すべき点がある（軽微） | IMPROVE | - | 問題なし | APPROVE | - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 問題なし | `[ARCHITECT:APPROVE]` | - | 軽微な改善必要 | `[ARCHITECT:IMPROVE]` | - | 構造的な修正必要 | `[ARCHITECT:REJECT]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: architect_review (アーキテクチャレビュー) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/03-architect-review.md + AI特有の問題についてコードをレビューしてください: + - 仮定の検証 + - もっともらしいが間違っているパターン + - 既存コードベースとの適合性 + - スコープクリープの検出 + rules: + - condition: AI特有の問題が見つからない + next: architect_review + - condition: AI特有の問題が検出された + next: ai_fix - ## Original User Request (ワークフロー開始時の元の要求) - {task} + - name: ai_fix + edit: true + agent: ~/.takt/agents/default/coder.md + allowed_tools: + - Read + - Glob + - Grep + - Edit + - Write + - Bash + - WebSearch + - WebFetch + instruction_template: | + ## AI Review Feedback (これが最新の指示です - 優先して対応してください) + {previous_response} - ## Git Diff - ```diff - {git_diff} - ``` + **重要**: AI Reviewerのフィードバックに対応してください。 + 以下に集中してください: + - 間違った仮定の修正 + - もっともらしいが間違っている実装の修正 + - 既存コードベースのパターンとの整合 + - スコープクリープの除去 + pass_previous_response: true + rules: + - condition: AI Reviewerの指摘に対する修正が完了した + next: ai_review + - condition: 修正を進行できない + next: plan - ## Instructions + # =========================================== + # Phase 3: Architecture Review + # =========================================== + - name: architect_review + edit: false + agent: ~/.takt/agents/default/architecture-reviewer.md + report: + name: 04-architect-review.md + format: | + ```markdown + # アーキテクチャレビュー + + ## 結果: APPROVE / IMPROVE / REJECT + + ## サマリー + {1-2文で結果を要約} + + ## 確認した観点 + - [x] 構造・設計 + - [x] コード品質 + - [x] 変更スコープ + - [x] テストカバレッジ + - [x] デッドコード + - [x] 呼び出しチェーン検証 + + ## 問題点（REJECTの場合） + | # | 場所 | 問題 | 修正案 | + |---|------|------|--------| + | 1 | `src/file.ts:42` | 問題の説明 | 修正方法 | + + ## 改善提案（任意・ブロッキングではない） + - {将来的な改善提案} + ``` + + **認知負荷軽減ルール:** + - APPROVE + 問題なし → サマリーのみ（5行以内） + - APPROVE + 軽微な提案 → サマリー + 改善提案（15行以内） + - REJECT → 問題点を表形式で（30行以内） + allowed_tools: + - Read + - Glob + - Grep + - Write + - WebSearch + - WebFetch + instruction_template: | **アーキテクチャと設計**のレビューに集中してください。 **レビュー観点:** @@ -254,50 +278,16 @@ steps: - テストカバレッジ - デッドコード - 呼び出しチェーン検証 - - **レポート出力:** 上記の `Report File` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - - **レポートフォーマット:** - ```markdown - # アーキテクチャレビュー - - ## 結果: APPROVE / IMPROVE / REJECT - - ## サマリー - {1-2文で結果を要約} - - ## 確認した観点 - - [x] 構造・設計 - - [x] コード品質 - - [x] 変更スコープ - - [x] テストカバレッジ - - [x] デッドコード - - [x] 呼び出しチェーン検証 - - ## 問題点（REJECTの場合） - | # | 場所 | 問題 | 修正案 | - |---|------|------|--------| - | 1 | `src/file.ts:42` | 問題の説明 | 修正方法 | - - ## 改善提案（任意・ブロッキングではない） - - {将来的な改善提案} - ``` - - **認知負荷軽減ルール:** - - APPROVE + 問題なし → サマリーのみ（5行以内） - - APPROVE + 軽微な提案 → サマリー + 改善提案（15行以内） - - REJECT → 問題点を表形式で（30行以内） - transitions: - - condition: approved - next_step: frontend_review - - condition: improve - next_step: fix_architect - - condition: rejected - next_step: fix_architect + rules: + - condition: アーキテクチャと設計に問題がない + next: frontend_review + - condition: 軽微な改善が必要だが構造的な問題はない + next: fix_architect + - condition: 構造的な問題があり修正が必要 + next: fix_architect - name: fix_architect + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -309,49 +299,51 @@ steps: - WebSearch - WebFetch permission_mode: acceptEdits - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 修正完了 | `[CODER:DONE]` | - | 進行不可 | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: fix_architect - ## Architect Feedback (これが最新の指示です - 優先して対応してください) {previous_response} - ## Original User Request (ワークフロー開始時の元の要求 - 参考情報) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **重要**: Architectのフィードバックに対応してください。 「Original User Request」は参考情報であり、最新の指示ではありません。 セッションの会話履歴を確認し、Architectの指摘事項を修正してください。 - - 完了時は [CODER:DONE] を含めてください。 - 進行できない場合は [CODER:BLOCKED] を含めてください。 pass_previous_response: true - transitions: - - condition: done - next_step: architect_review - - condition: blocked - next_step: plan + rules: + - condition: Architectの指摘に対する修正が完了した + next: architect_review + - condition: 修正を進行できない + next: plan # =========================================== - # Phase 3: Frontend Review + # Phase 4: Frontend Review # =========================================== - name: frontend_review + edit: false agent: ~/.takt/agents/expert/frontend-reviewer.md + report: + name: 05-frontend-review.md + format: | + ```markdown + # フロントエンドレビュー + + ## 結果: APPROVE / REJECT + + ## サマリー + {1-2文で結果を要約} + + ## 確認した観点 + | 観点 | 結果 | 備考 | + |------|------|------| + | コンポーネント設計 | ✅ | - | + | 状態管理 | ✅ | - | + | パフォーマンス | ✅ | - | + | アクセシビリティ | ✅ | - | + | 型安全性 | ✅ | - | + + ## 問題点（REJECTの場合） + | # | 場所 | 問題 | 修正案 | + |---|------|------|--------| + | 1 | `src/file.tsx:42` | 問題の説明 | 修正方法 | + ``` allowed_tools: - Read - Glob @@ -359,32 +351,7 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | フロントエンド設計に問題なし | `[FRONTEND:APPROVE]` | - | フロントエンド設計に問題あり | `[FRONTEND:REJECT]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: frontend_review (フロントエンド専門レビュー) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/04-frontend-review.md - - ## Original User Request - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions フロントエンド開発の観点から上記の変更をレビューしてください。 **レビュー観点:** @@ -396,46 +363,15 @@ steps: - TypeScript型安全性 **注意**: このプロジェクトがフロントエンドを含まない場合は、 - [FRONTEND:APPROVE] として次に進んでください。 - - **レポート出力:** 上記の `Report File` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - - **レポートフォーマット:** - ```markdown - # フロントエンドレビュー - - ## 結果: APPROVE / REJECT - - ## サマリー - {1-2文で結果を要約} - - ## 確認した観点 - | 観点 | 結果 | 備考 | - |------|------|------| - | コンポーネント設計 | ✅ | - | - | 状態管理 | ✅ | - | - | パフォーマンス | ✅ | - | - | アクセシビリティ | ✅ | - | - | 型安全性 | ✅ | - | - - ## 問題点（REJECTの場合） - | # | 場所 | 問題 | 修正案 | - |---|------|------|--------| - | 1 | `src/file.tsx:42` | 問題の説明 | 修正方法 | - ``` - - Include: - - [FRONTEND:APPROVE] if frontend design is sound - - [FRONTEND:REJECT] if design issues found (list specific issues) - transitions: - - condition: approved - next_step: ai_review - - condition: rejected - next_step: fix_frontend + 問題なしとして次に進んでください。 + rules: + - condition: フロントエンド設計に問題がない + next: security_review + - condition: フロントエンド設計に問題がある + next: fix_frontend - name: fix_frontend + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -446,31 +382,10 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 修正完了 | `[CODER:DONE]` | - | 進行不可 | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: fix_frontend - ## Frontend Review Feedback (これが最新の指示です - 優先して対応してください) {previous_response} - ## Original User Request (ワークフロー開始時の元の要求 - 参考情報) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **重要**: フロントエンド専門家からの指摘を修正してください。 指摘されたポイント: @@ -479,157 +394,45 @@ steps: - パフォーマンス - アクセシビリティ - 型安全性 - - 完了時は [CODER:DONE] を含めてください。 - 進行できない場合は [CODER:BLOCKED] を含めてください。 pass_previous_response: true - transitions: - - condition: done - next_step: frontend_review - - condition: blocked - next_step: plan - - # =========================================== - # Phase 4: AI Review - # =========================================== - - name: ai_review - agent: ~/.takt/agents/default/ai-antipattern-reviewer.md - allowed_tools: - - Read - - Glob - - Grep - - Write - - WebSearch - - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | AI特有の問題なし | `[AI_REVIEW:APPROVE]` | - | 問題あり | `[AI_REVIEW:REJECT]` | - instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: ai_review (AI生成コードレビュー) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/05-ai-review.md - - ## Original User Request (ワークフロー開始時の元の要求) - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions - AI特有の問題についてコードをレビューしてください: - - 仮定の検証 - - もっともらしいが間違っているパターン - - 既存コードベースとの適合性 - - スコープクリープの検出 - - **レポート出力:** 上記の `Report File` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - - **レポートフォーマット:** - ```markdown - # AI生成コードレビュー - - ## 結果: APPROVE / REJECT - - ## サマリー - {1文で結果を要約} - - ## 検証した項目 - | 観点 | 結果 | 備考 | - |------|------|------| - | 仮定の妥当性 | ✅ | - | - | API/ライブラリの実在 | ✅ | - | - | コンテキスト適合 | ✅ | - | - | スコープ | ✅ | - | - - ## 問題点（REJECTの場合） - | # | カテゴリ | 場所 | 問題 | - |---|---------|------|------| - | 1 | 幻覚API | `src/file.ts:23` | 存在しないメソッド | - ``` - - **認知負荷軽減ルール:** - - 問題なし → サマリー1文 + チェック表のみ（10行以内） - - 問題あり → + 問題を表形式で（25行以内） - - 以下を含めてください: - - [AI_REVIEW:APPROVE] AI特有の問題が見つからない場合 - - [AI_REVIEW:REJECT] 問題が検出された場合（具体的な問題をリスト） - transitions: - - condition: approved - next_step: security_review - - condition: rejected - next_step: ai_fix - - - name: ai_fix - agent: ~/.takt/agents/default/coder.md - allowed_tools: - - Read - - Glob - - Grep - - Edit - - Write - - Bash - - WebSearch - - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 修正完了 | `[CODER:DONE]` | - | 進行不可 | `[CODER:BLOCKED]` | - instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: ai_fix - - ## AI Review Feedback (これが最新の指示です - 優先して対応してください) - {previous_response} - - ## Original User Request (ワークフロー開始時の元の要求 - 参考情報) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions - **重要**: AI Reviewerのフィードバックに対応してください。 - 以下に集中してください: - - 間違った仮定の修正 - - もっともらしいが間違っている実装の修正 - - 既存コードベースのパターンとの整合 - - スコープクリープの除去 - - 完了時は [CODER:DONE] を含めてください。 - 進行できない場合は [CODER:BLOCKED] を含めてください。 - pass_previous_response: true - transitions: - - condition: done - next_step: ai_review - - condition: blocked - next_step: plan + rules: + - condition: フロントエンドの指摘に対する修正が完了した + next: frontend_review + - condition: 修正を進行できない + next: plan # =========================================== # Phase 5: Security Review # =========================================== - name: security_review + edit: false agent: ~/.takt/agents/expert/security-reviewer.md + report: + name: 06-security-review.md + format: | + ```markdown + # セキュリティレビュー + + ## 結果: APPROVE / REJECT + + ## 重大度: None / Low / Medium / High / Critical + + ## チェック結果 + | カテゴリ | 結果 | 備考 | + |---------|------|------| + | インジェクション | ✅ | - | + | 認証・認可 | ✅ | - | + | データ保護 | ✅ | - | + | 依存関係 | ✅ | - | + + ## 脆弱性（REJECTの場合） + | # | 重大度 | 種類 | 場所 | 修正案 | + |---|--------|------|------|--------| + | 1 | High | SQLi | `src/db.ts:42` | パラメータ化クエリを使用 | + + ## 警告（ブロッキングではない） + - {セキュリティに関する推奨事項} + ``` allowed_tools: - Read - Glob @@ -637,32 +440,7 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | セキュリティ問題なし | `[SECURITY:APPROVE]` | - | 脆弱性あり | `[SECURITY:REJECT]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: security_review (セキュリティ専門レビュー) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/06-security-review.md - - ## Original User Request - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions セキュリティの観点から上記の変更をレビューしてください。 **レビュー観点:** @@ -671,46 +449,14 @@ steps: - 機密情報の取り扱い - 暗号化の適切性 - OWASP Top 10 - - **レポート出力:** 上記の `Report File` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - - **レポートフォーマット:** - ```markdown - # セキュリティレビュー - - ## 結果: APPROVE / REJECT - - ## 重大度: None / Low / Medium / High / Critical - - ## チェック結果 - | カテゴリ | 結果 | 備考 | - |---------|------|------| - | インジェクション | ✅ | - | - | 認証・認可 | ✅ | - | - | データ保護 | ✅ | - | - | 依存関係 | ✅ | - | - - ## 脆弱性（REJECTの場合） - | # | 重大度 | 種類 | 場所 | 修正案 | - |---|--------|------|------|--------| - | 1 | High | SQLi | `src/db.ts:42` | パラメータ化クエリを使用 | - - ## 警告（ブロッキングではない） - - {セキュリティに関する推奨事項} - ``` - - Include: - - [SECURITY:APPROVE] if no security issues found - - [SECURITY:REJECT] if vulnerabilities found (list specific issues with severity) - transitions: - - condition: approved - next_step: qa_review - - condition: rejected - next_step: fix_security + rules: + - condition: セキュリティ上の問題がない + next: qa_review + - condition: セキュリティ上の脆弱性が検出された + next: fix_security - name: fix_security + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -721,32 +467,10 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 軽微な修正完了 | `[CODER:DONE]` | - | 大きな修正（アーキテクチャからやり直し） | `[CODER:REJECT]` | - | 進行不可 | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: fix_security - ## Security Review Feedback (これが最新の指示です - 優先して対応してください) {previous_response} - ## Original User Request (ワークフロー開始時の元の要求 - 参考情報) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **重要**: セキュリティ専門家からの指摘を修正してください。 セキュリティ問題は最優先で対応してください。 @@ -757,28 +481,52 @@ steps: - 暗号化の問題 ## 修正完了時の判断 - 修正が完了したら、**変更の影響範囲**を判断して適切なタグを出力してください： + 修正が完了したら、**変更の影響範囲**を判断してください： - - `[CODER:DONE]` - 軽微な修正（セキュリティレビューのみ再実施） + - 軽微な修正（セキュリティレビューのみ再実施） - 例: バリデーション追加、エスケープ処理追加、設定変更 - - `[CODER:REJECT]` - 大きな修正（アーキテクチャレビューからやり直し） + - 大きな修正（アーキテクチャレビューからやり直し） - 例: データフロー変更、API設計変更、認証方式変更、ドメインモデル変更 - - 進行できない場合は [CODER:BLOCKED] を含めてください。 pass_previous_response: true - transitions: - - condition: done - next_step: security_review - - condition: rejected - next_step: architect_review - - condition: blocked - next_step: plan + rules: + - condition: 軽微なセキュリティ修正が完了した + next: security_review + - condition: 大きな修正を行い、アーキテクチャレビューからやり直す必要がある + next: architect_review + - condition: 修正を進行できない + next: plan # =========================================== # Phase 6: QA Review # =========================================== - name: qa_review + edit: false agent: ~/.takt/agents/expert/qa-reviewer.md + report: + name: 07-qa-review.md + format: | + ```markdown + # QAレビュー + + ## 結果: APPROVE / REJECT + + ## サマリー + {1-2文で結果を要約} + + ## 確認した観点 + | 観点 | 結果 | 備考 | + |------|------|------| + | テストカバレッジ | ✅ | - | + | テスト品質 | ✅ | - | + | エラーハンドリング | ✅ | - | + | ドキュメント | ✅ | - | + | 保守性 | ✅ | - | + + ## 問題点（REJECTの場合） + | # | カテゴリ | 問題 | 修正案 | + |---|---------|------|--------| + | 1 | テスト | 問題の説明 | 修正方法 | + ``` allowed_tools: - Read - Glob @@ -786,32 +534,7 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 品質基準を満たす | `[QA:APPROVE]` | - | 品質問題あり | `[QA:REJECT]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: qa_review (QA専門レビュー) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/07-qa-review.md - - ## Original User Request - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions 品質保証の観点から上記の変更をレビューしてください。 **レビュー観点:** @@ -821,45 +544,14 @@ steps: - エラーハンドリング - ログとモニタリング - 保守性 - - **レポート出力:** 上記の `Report File` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - - **レポートフォーマット:** - ```markdown - # QAレビュー - - ## 結果: APPROVE / REJECT - - ## サマリー - {1-2文で結果を要約} - - ## 確認した観点 - | 観点 | 結果 | 備考 | - |------|------|------| - | テストカバレッジ | ✅ | - | - | テスト品質 | ✅ | - | - | エラーハンドリング | ✅ | - | - | ドキュメント | ✅ | - | - | 保守性 | ✅ | - | - - ## 問題点（REJECTの場合） - | # | カテゴリ | 問題 | 修正案 | - |---|---------|------|--------| - | 1 | テスト | 問題の説明 | 修正方法 | - ``` - - Include: - - [QA:APPROVE] if quality standards are met - - [QA:REJECT] if quality issues found (list specific issues) - transitions: - - condition: approved - next_step: supervise - - condition: rejected - next_step: fix_qa + rules: + - condition: 品質基準を満たしている + next: supervise + - condition: 品質に問題がある + next: fix_qa - name: fix_qa + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -870,33 +562,10 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 軽微な修正完了 | `[CODER:DONE]` | - | セキュリティ影響あり | `[CODER:IMPROVE]` | - | 大きな修正（アーキテクチャからやり直し） | `[CODER:REJECT]` | - | 進行不可 | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: fix_qa - ## QA Review Feedback (これが最新の指示です - 優先して対応してください) {previous_response} - ## Original User Request (ワークフロー開始時の元の要求 - 参考情報) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **重要**: QA専門家からの指摘を修正してください。 指摘されたポイント: @@ -907,32 +576,34 @@ steps: - コード品質 ## 修正完了時の判断 - 修正が完了したら、**変更の影響範囲**を判断して適切なタグを出力してください： + 修正が完了したら、**変更の影響範囲**を判断してください： - - `[CODER:DONE]` - 軽微な修正（QAレビューのみ再実施） + - 軽微な修正（QAレビューのみ再実施） - 例: テスト追加、ドキュメント追加、ログ追加、コメント追加 - - `[CODER:IMPROVE]` - セキュリティに影響する修正（セキュリティレビューからやり直し） + - セキュリティに影響する修正（セキュリティレビューからやり直し） - 例: エラーハンドリング変更（エラーメッセージの内容変更）、入力検証の変更 - - `[CODER:REJECT]` - 大きな修正（アーキテクチャレビューからやり直し） + - 大きな修正（アーキテクチャレビューからやり直し） - 例: ビジネスロジック変更、データモデル変更、API変更 - - 進行できない場合は [CODER:BLOCKED] を含めてください。 pass_previous_response: true - transitions: - - condition: done - next_step: qa_review - - condition: improve - next_step: security_review - - condition: rejected - next_step: architect_review - - condition: blocked - next_step: plan + rules: + - condition: 軽微なQA修正が完了した + next: qa_review + - condition: セキュリティに影響する修正を行った + next: security_review + - condition: 大きな修正を行い、アーキテクチャレビューからやり直す必要がある + next: architect_review + - condition: 修正を進行できない + next: plan # =========================================== # Phase 7: Supervision # =========================================== - name: supervise + edit: false agent: ~/.takt/agents/expert/supervisor.md + report: + - Validation: 08-supervisor-validation.md + - Summary: summary.md allowed_tools: - Read - Glob @@ -940,42 +611,15 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 最終承認 | `[SUPERVISOR:APPROVE]` | - | 問題あり | `[SUPERVISOR:REJECT]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: supervise (最終確認) - - Report Directory: .takt/reports/{report_dir}/ - - Report Files: - - Validation: .takt/reports/{report_dir}/08-supervisor-validation.md - - Summary: .takt/reports/{report_dir}/summary.md - - ## Original User Request - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - ## Previous Reviews Summary このステップに到達したということは、以下のレビューがすべてAPPROVEされています： + - AI Review: APPROVED - Architecture Review: APPROVED - Frontend Review: APPROVED - - AI Review: APPROVED - Security Review: APPROVED - QA Review: APPROVED - ## Instructions テスト実行、ビルド確認、最終承認を行ってください。 **ワークフロー全体の確認:** @@ -986,10 +630,6 @@ steps: **レポートの確認:** Report Directory内の全レポートを読み、 未対応の改善提案がないか確認してください。 - **レポート出力:** 上記の `Report Files` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - **Validationレポートフォーマット:** ```markdown # 最終検証結果 @@ -1032,9 +672,9 @@ steps: ## レビュー結果 | レビュー | 結果 | |---------|------| + | AI Review | ✅ APPROVE | | Architecture | ✅ APPROVE | | Frontend | ✅ APPROVE | - | AI Review | ✅ APPROVE | | Security | ✅ APPROVE | | QA | ✅ APPROVE | | Supervisor | ✅ APPROVE | @@ -1045,17 +685,14 @@ steps: npm run build ``` ``` - - 出力: - - [SUPERVISOR:APPROVE] すべて完了、マージ可能 - - [SUPERVISOR:REJECT] 問題あり（具体的な問題を記載） - transitions: - - condition: approved - next_step: COMPLETE - - condition: rejected - next_step: fix_supervisor + rules: + - condition: すべての検証が完了し、マージ可能な状態である + next: COMPLETE + - condition: 問題が検出された + next: fix_supervisor - name: fix_supervisor + edit: true agent: ~/.takt/agents/default/coder.md allowed_tools: - Read @@ -1066,41 +703,17 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 修正完了 | `[CODER:DONE]` | - | 進行不可 | `[CODER:BLOCKED]` | instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: fix_supervisor - ## Supervisor Feedback (これが最新の指示です - 優先して対応してください) {previous_response} - ## Original User Request (ワークフロー開始時の元の要求 - 参考情報) - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions **重要**: 監督者からの指摘を修正してください。 監督者は全体を俯瞰した視点から問題を指摘しています。 優先度の高い項目から順に対応してください。 - - 完了時は [CODER:DONE] を含めてください。 - 進行できない場合は [CODER:BLOCKED] を含めてください。 pass_previous_response: true - transitions: - - condition: done - next_step: supervise - - condition: blocked - next_step: plan + rules: + - condition: 監督者の指摘に対する修正が完了した + next: supervise + - condition: 修正を進行できない + next: plan diff --git a/resources/global/ja/workflows/magi.yaml b/resources/global/ja/workflows/magi.yaml index 32c77dc..882758a 100644 --- a/resources/global/ja/workflows/magi.yaml +++ b/resources/global/ja/workflows/magi.yaml @@ -8,7 +8,6 @@ # {step_iteration} - ステップごとのイテレーション数（このステップが何回実行されたか） # {task} - 元のユーザー要求 # {previous_response} - 前のステップの出力 -# {git_diff} - 現在のコミットされていない変更（git diff） # {user_inputs} - ワークフロー中に蓄積されたユーザー入力 # {report_dir} - レポートディレクトリ名（例: "20250126-143052-task-summary"） @@ -26,8 +25,15 @@ steps: - Grep - WebSearch - WebFetch - status_rules_prompt: | + instruction_template: | + # MAGI System 起動 + ## 審議事項 + {task} + + ## 指示 + あなたはMAGI System の MELCHIOR-1 です。 + 科学者・技術者の観点から上記を分析し、判定を下してください。 ## 出力フォーマット @@ -44,18 +50,9 @@ steps: 理由: {賛成の理由} ``` - instruction_template: | - # MAGI System 起動 - - ## 審議事項 - {task} - - ## 指示 - あなたはMAGI System の MELCHIOR-1 です。 - 科学者・技術者の観点から上記を分析し、判定を下してください。 - transitions: - - condition: always - next_step: balthasar + rules: + - condition: 判定を完了した + next: balthasar - name: balthasar agent: ~/.takt/agents/magi/balthasar.md @@ -65,8 +62,19 @@ steps: - Grep - WebSearch - WebFetch - status_rules_prompt: | + instruction_template: | + # MAGI System 継続 + ## 審議事項 + {task} + + ## MELCHIOR-1 の判定 + {previous_response} + + ## 指示 + あなたはMAGI System の BALTHASAR-2 です。 + 育成者の観点から上記を分析し、判定を下してください。 + MELCHIORの判定は参考にしつつも、独自の観点で判断してください。 ## 出力フォーマット @@ -83,23 +91,10 @@ steps: 理由: {賛成の理由} ``` - instruction_template: | - # MAGI System 継続 - - ## 審議事項 - {task} - - ## MELCHIOR-1 の判定 - {previous_response} - - ## 指示 - あなたはMAGI System の BALTHASAR-2 です。 - 育成者の観点から上記を分析し、判定を下してください。 - MELCHIORの判定は参考にしつつも、独自の観点で判断してください。 pass_previous_response: true - transitions: - - condition: always - next_step: casper + rules: + - condition: 判定を完了した + next: casper - name: casper agent: ~/.takt/agents/magi/casper.md @@ -109,8 +104,20 @@ steps: - Grep - WebSearch - WebFetch - status_rules_prompt: | + instruction_template: | + # MAGI System 最終審議 + ## 審議事項 + {task} + + ## これまでの判定 + {previous_response} + + ## 指示 + あなたはMAGI System の CASPER-3 です。 + 実務・現実の観点から上記を分析し、判定を下してください。 + + **最後に、3者の判定を集計し、最終結論を出してください。** ## 出力フォーマット @@ -137,21 +144,7 @@ steps: [理由・まとめ] ``` - instruction_template: | - # MAGI System 最終審議 - - ## 審議事項 - {task} - - ## これまでの判定 - {previous_response} - - ## 指示 - あなたはMAGI System の CASPER-3 です。 - 実務・現実の観点から上記を分析し、判定を下してください。 - - **最後に、3者の判定を集計し、最終結論を出してください。** pass_previous_response: true - transitions: - - condition: always - next_step: COMPLETE + rules: + - condition: 最終判定を完了した + next: COMPLETE diff --git a/resources/global/ja/workflows/research.yaml b/resources/global/ja/workflows/research.yaml index 0f7fce7..a82d640 100644 --- a/resources/global/ja/workflows/research.yaml +++ b/resources/global/ja/workflows/research.yaml @@ -12,7 +12,6 @@ # {step_iteration} - ステップごとのイテレーション数（このステップが何回実行されたか） # {task} - 元のユーザー要求 # {previous_response} - 前のステップの出力 -# {git_diff} - 現在のコミットされていない変更（git diff） # {user_inputs} - ワークフロー中に蓄積されたユーザー入力 # {report_dir} - レポートディレクトリ名（例: "20250126-143052-task-summary"） @@ -30,30 +29,6 @@ steps: - Grep - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 計画完了 | `[PLANNER:DONE]` | - | 情報不足 | `[PLANNER:BLOCKED]` | - - ### 出力例 - - **DONE の場合:** - ``` - [PLANNER:DONE] - ``` - - **BLOCKED の場合:** - ``` - [PLANNER:BLOCKED] - - 確認事項: - - {質問1} - ``` instruction_template: | ## ワークフロー状況 - イテレーション: {iteration}/{max_iterations}（ワークフロー全体） @@ -77,11 +52,11 @@ steps: - 複数の解釈がある場合は、すべてを調査対象に含める - Supervisorからフィードバックがある場合は、指摘を反映した計画を作成 pass_previous_response: true - transitions: - - condition: done - next_step: dig - - condition: blocked - next_step: ABORT + rules: + - condition: 計画が完了した + next: dig + - condition: 情報が不足しており計画を立てられない + next: ABORT - name: dig agent: ~/.takt/agents/research/digger.md @@ -91,29 +66,6 @@ steps: - Grep - WebSearch - WebFetch - status_rules_prompt: | - - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 調査完了 | `[DIGGER:DONE]` | - | 調査不可 | `[DIGGER:BLOCKED]` | - - ### 出力例 - - **DONE の場合:** - ``` - [DIGGER:DONE] - ``` - - **BLOCKED の場合:** - ``` - [DIGGER:BLOCKED] - - 理由: {調査できなかった理由} - ``` instruction_template: | ## ワークフロー状況 - イテレーション: {iteration}/{max_iterations}（ワークフロー全体） @@ -142,11 +94,11 @@ steps: - コードベース検索 - ファイル読み取り pass_previous_response: true - transitions: - - condition: done - next_step: supervise - - condition: blocked - next_step: ABORT + rules: + - condition: 調査が完了した + next: supervise + - condition: 調査を実行できない + next: ABORT - name: supervise agent: ~/.takt/agents/research/supervisor.md @@ -156,39 +108,6 @@ steps: - Grep - WebSearch - WebFetch - status_rules_prompt: | - - - ## 判定基準 - - | 状況 | 判定 | - |------|------| - | 調査結果が十分 | APPROVE | - | 調査結果が不十分 | REJECT | - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 調査完了、結果は十分 | `[SUPERVISOR:APPROVE]` | - | 不十分、計画からやり直し | `[SUPERVISOR:REJECT]` | - - ### 出力例 - - **APPROVE の場合:** - ``` - [SUPERVISOR:APPROVE] - - 調査結果は元の依頼に対して十分です。 - ``` - - **REJECT の場合:** - ``` - [SUPERVISOR:REJECT] - - 不足点: - - {具体的な不足点} - ``` instruction_template: | ## ワークフロー状況 - イテレーション: {iteration}/{max_iterations}（ワークフロー全体） @@ -206,10 +125,10 @@ steps: **重要**: 問題がある場合は、Plannerへの具体的な指示を含めてください。 pass_previous_response: true - transitions: - - condition: approved - next_step: COMPLETE - - condition: rejected - next_step: plan + rules: + - condition: 調査結果が元の依頼に対して十分である + next: COMPLETE + - condition: 調査結果が不十分であり、計画からやり直す必要がある + next: plan initial_step: plan diff --git a/resources/global/ja/workflows/simple.yaml b/resources/global/ja/workflows/simple.yaml index 4dfab57..fd86d39 100644 --- a/resources/global/ja/workflows/simple.yaml +++ b/resources/global/ja/workflows/simple.yaml @@ -1,19 +1,18 @@ # Simple TAKT Workflow -# Plan -> Coder -> Architect Review -> AI Review -> Supervisor Approval +# Plan -> Implement -> AI Review -> Architect Review -> Supervisor Approval # (defaultの簡略版 - improve, fix, ai_fix, security_review, security_fix を削除) # -# Template Variables: +# Template Variables (auto-injected): # {iteration} - Workflow-wide turn count (total steps executed across all agents) # {max_iterations} - Maximum iterations allowed for the workflow # {step_iteration} - Per-step iteration count (how many times THIS step has been executed) -# {task} - Original user request -# {previous_response} - Output from the previous step -# {git_diff} - Current uncommitted changes (git diff) -# {user_inputs} - Accumulated user inputs during workflow +# {task} - Original user request (auto-injected) +# {previous_response} - Output from the previous step (auto-injected) +# {user_inputs} - Accumulated user inputs during workflow (auto-injected) # {report_dir} - Report directory name (e.g., "20250126-143052-task-summary") name: simple -description: Simplified development workflow (plan -> implement -> review -> ai_review -> supervise) +description: Simplified development workflow (plan -> implement -> ai_review -> review -> supervise) max_iterations: 20 @@ -21,7 +20,31 @@ initial_step: plan steps: - name: plan + edit: false agent: ~/.takt/agents/default/planner.md + report: + name: 00-plan.md + format: | + ```markdown + # タスク計画 + + ## 元の要求 + {ユーザーの要求をそのまま記載} + + ## 分析結果 + + ### 目的 + {達成すべきこと} + + ### スコープ + {影響範囲} + + ### 実装アプローチ + {どう進めるか} + + ## 確認事項（あれば） + - {不明点や確認が必要な点} + ``` allowed_tools: - Read - Glob @@ -30,67 +53,12 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - ## 判定基準 - - | 状況 | 判定 | - |------|------| - | 要件が明確で実装可能 | DONE | - | ユーザーが質問をしている（実装タスクではない） | ANSWER | - | 要件が不明確、情報不足 | BLOCKED | - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 分析完了 | `[PLANNER:DONE]` | - | 質問への回答 | `[PLANNER:ANSWER]` | - | 情報不足 | `[PLANNER:BLOCKED]` | - - ### 出力例 - - **DONE の場合:** - ``` - [PLANNER:DONE] - ``` - - **ANSWER の場合:** - ``` - {質問への回答} - - [PLANNER:ANSWER] - ``` - - **BLOCKED の場合:** - ``` - [PLANNER:BLOCKED] - - 確認事項: - - {質問1} - - {質問2} - ``` instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: plan (タスク分析) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/00-plan.md - - ## User Request - {task} - ## Previous Response (implementからの差し戻し時) {previous_response} - ## Instructions タスクを分析し、実装方針を立ててください。 - **判断基準:** - - ユーザーの入力が実装タスクの場合 → 計画を立てて `[PLANNER:DONE]` - - ユーザーの入力が質問の場合 → 調査・回答して `[PLANNER:ANSWER]` - - 情報不足の場合 → `[PLANNER:BLOCKED]` - **注意:** Previous Responseがある場合は差し戻しのため、 その内容を踏まえて計画を見直してください（replan）。 @@ -98,43 +66,25 @@ steps: 1. タスクの要件を理解する 2. 影響範囲を特定する 3. 実装アプローチを決める - - **レポート出力:** 上記の `Report File` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - - **レポートフォーマット:** - ```markdown - # タスク計画 - - ## 元の要求 - {ユーザーの要求をそのまま記載} - - ## 分析結果 - - ### 目的 - {達成すべきこと} - - ### スコープ - {影響範囲} - - ### 実装アプローチ - {どう進めるか} - - ## 確認事項（あれば） - - {不明点や確認が必要な点} - ``` pass_previous_response: true - transitions: - - condition: done - next_step: implement - - condition: answer - next_step: COMPLETE - - condition: blocked - next_step: ABORT + rules: + - condition: "要件が明確で実装可能" + next: implement + - condition: "ユーザーが質問をしている（実装タスクではない）" + next: COMPLETE + - condition: "要件が不明確、情報不足" + next: ABORT + appendix: | + 確認事項: + - {質問1} + - {質問2} - name: implement + edit: true agent: ~/.takt/agents/default/coder.md + report: + - Scope: 01-coder-scope.md + - Decisions: 02-coder-decisions.md allowed_tools: - Read - Glob @@ -145,58 +95,10 @@ steps: - WebSearch - WebFetch permission_mode: acceptEdits - status_rules_prompt: | - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 実装完了 | `[CODER:DONE]` | - | 判断できない/情報不足 | `[CODER:BLOCKED]` | - - **重要**: 迷ったら `[BLOCKED]`。勝手に判断しない。 - - ### 出力例 - - **DONE の場合:** - ``` - 実装完了しました。 - - 作成: `src/auth/service.ts`, `tests/auth.test.ts` - - 変更: `src/routes.ts` - - [CODER:DONE] - ``` - - **BLOCKED の場合:** - ``` - [CODER:BLOCKED] - - 理由: DBスキーマが未定義のため実装できません - 必要な情報: usersテーブルの構造 - ``` instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: implement - - Report Directory: .takt/reports/{report_dir}/ - - Report Files: - - Scope: .takt/reports/{report_dir}/01-coder-scope.md - - Decisions: .takt/reports/{report_dir}/02-coder-decisions.md - - ## User Request - {task} - - ## Additional User Inputs - {user_inputs} - - ## Instructions planステップで立てた計画に従って実装してください。 計画レポート（00-plan.md）を参照し、実装を進めてください。 - **レポート出力:** 上記の `Report Files` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - **Scopeレポートフォーマット（実装開始時に作成）:** ```markdown # 変更スコープ宣言 @@ -226,125 +128,43 @@ steps: - **検討した選択肢**: {選択肢リスト} - **理由**: {選んだ理由} ``` - transitions: - - condition: done - next_step: review - - condition: blocked - next_step: plan - - - name: review - agent: ~/.takt/agents/default/architect.md - allowed_tools: - - Read - - Glob - - Grep - - Write - - WebSearch - - WebFetch - status_rules_prompt: | - ## 判定基準 - - | 状況 | 判定 | - |------|------| - | 構造に問題がある | REJECT | - | 設計原則違反がある | REJECT | - | セキュリティ問題がある | REJECT | - | テストが不十分 | REJECT | - | 問題なし | APPROVE | - - **注意:** simpleワークフローでは IMPROVE は使用しません。 - 軽微な問題もある場合は APPROVE + コメントとしてください。 - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 問題なし | `[ARCHITECT:APPROVE]` | - | 構造的な修正必要 | `[ARCHITECT:REJECT]` | - - ### 出力例 - - **APPROVE の場合:** - ``` - [ARCHITECT:APPROVE] - - 良い点: - - モジュール分割が適切 - - 単一責務が守られている - ``` - - **REJECT の場合:** - ``` - [ARCHITECT:REJECT] - - 問題点: - 1. ファイルサイズ超過 - - 場所: `src/services/user.ts` (523行) - - 修正案: 3ファイルに分割 - ``` - instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: review (アーキテクチャレビュー) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/03-architect-review.md - - ## Original User Request (ワークフロー開始時の元の要求) - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions - **アーキテクチャと設計**のレビューに集中してください。AI特有の問題はレビューしないでください（次のステップで行います）。 - - 変更をレビューしてフィードバックを提供してください。 - - **注意:** simpleワークフローではIMPROVE判定は使用しません。 - 軽微な改善提案がある場合は APPROVE + コメントとしてください。 - - **レポート出力:** 上記の `Report File` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - - **レポートフォーマット:** - ```markdown - # アーキテクチャレビュー - - ## 結果: APPROVE / REJECT - - ## サマリー - {1-2文で結果を要約} - - ## 確認した観点 - - [x] 構造・設計 - - [x] コード品質 - - [x] 変更スコープ - - ## 問題点（REJECTの場合） - | # | 場所 | 問題 | 修正案 | - |---|------|------|--------| - | 1 | `src/file.ts:42` | 問題の説明 | 修正方法 | - - ## 改善提案（任意・ブロッキングではない） - - {将来的な改善提案} - ``` - - **認知負荷軽減ルール:** - - APPROVE + 問題なし → サマリーのみ（5行以内） - - APPROVE + 軽微な提案 → サマリー + 改善提案（15行以内） - - REJECT → 問題点を表形式で（30行以内） - transitions: - - condition: approved - next_step: ai_review - - condition: rejected - next_step: plan + rules: + - condition: "実装完了" + next: ai_review + - condition: "判断できない、情報不足" + next: plan - name: ai_review + edit: false agent: ~/.takt/agents/default/ai-antipattern-reviewer.md + report: + name: 03-ai-review.md + format: | + ```markdown + # AI生成コードレビュー + + ## 結果: APPROVE / REJECT + + ## サマリー + {1文で結果を要約} + + ## 検証した項目 + | 観点 | 結果 | 備考 | + |------|------|------| + | 仮定の妥当性 | ✅ | - | + | API/ライブラリの実在 | ✅ | - | + | コンテキスト適合 | ✅ | - | + | スコープ | ✅ | - | + + ## 問題点（REJECTの場合） + | # | カテゴリ | 場所 | 問題 | + |---|---------|------|------| + | 1 | 幻覚API | `src/file.ts:23` | 存在しないメソッド | + ``` + + **認知負荷軽減ルール:** + - 問題なし → サマリー1文 + チェック表のみ（10行以内） + - 問題あり → + 問題を表形式で（25行以内） allowed_tools: - Read - Glob @@ -352,104 +172,76 @@ steps: - Write - WebSearch - WebFetch - status_rules_prompt: | - ## 判定基準 - - | 状況 | 判定 | - |------|------| - | 仮定が間違っている（動作に影響） | REJECT | - | もっともらしいが間違っているコード | REJECT | - | コードベースの文脈に重大な不整合 | REJECT | - | スコープクリープ | APPROVE（警告を付記） | - | 軽微なスタイルの逸脱のみ | APPROVE | - | コードが文脈に合い動作する | APPROVE | - - **注意:** スコープクリープは警告として記載するが、それだけでREJECTしない。 - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | AI特有の問題なし | `[AI_REVIEW:APPROVE]` | - | 問題あり | `[AI_REVIEW:REJECT]` | - - ### 出力例 - - **APPROVE の場合:** - ``` - [AI_REVIEW:APPROVE] - - 検証結果: 問題なし - ``` - - **REJECT の場合:** - ``` - [AI_REVIEW:REJECT] - - 問題点: - 1. 存在しないAPIを使用: `fetch.json()` → `response.json()` - ``` instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: ai_review (AI生成コードレビュー) - - Report Directory: .takt/reports/{report_dir}/ - - Report File: .takt/reports/{report_dir}/04-ai-review.md - - ## Original User Request (ワークフロー開始時の元の要求) - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions AI特有の問題についてコードをレビューしてください: - 仮定の検証 - もっともらしいが間違っているパターン - 既存コードベースとの適合性 - スコープクリープの検出 + rules: + - condition: "AI特有の問題なし" + next: review + - condition: "AI特有の問題あり" + next: plan - **レポート出力:** 上記の `Report File` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 + - name: review + edit: false + agent: ~/.takt/agents/default/architecture-reviewer.md + report: + name: 04-architect-review.md + format: | + ```markdown + # アーキテクチャレビュー - **レポートフォーマット:** - ```markdown - # AI生成コードレビュー + ## 結果: APPROVE / REJECT - ## 結果: APPROVE / REJECT + ## サマリー + {1-2文で結果を要約} - ## サマリー - {1文で結果を要約} + ## 確認した観点 + - [x] 構造・設計 + - [x] コード品質 + - [x] 変更スコープ - ## 検証した項目 - | 観点 | 結果 | 備考 | - |------|------|------| - | 仮定の妥当性 | ✅ | - | - | API/ライブラリの実在 | ✅ | - | - | コンテキスト適合 | ✅ | - | - | スコープ | ✅ | - | + ## 問題点（REJECTの場合） + | # | 場所 | 問題 | 修正案 | + |---|------|------|--------| + | 1 | `src/file.ts:42` | 問題の説明 | 修正方法 | - ## 問題点（REJECTの場合） - | # | カテゴリ | 場所 | 問題 | - |---|---------|------|------| - | 1 | 幻覚API | `src/file.ts:23` | 存在しないメソッド | - ``` + ## 改善提案（任意・ブロッキングではない） + - {将来的な改善提案} + ``` - **認知負荷軽減ルール:** - - 問題なし → サマリー1文 + チェック表のみ（10行以内） - - 問題あり → + 問題を表形式で（25行以内） - transitions: - - condition: approved - next_step: supervise - - condition: rejected - next_step: plan + **認知負荷軽減ルール:** + - APPROVE + 問題なし → サマリーのみ（5行以内） + - APPROVE + 軽微な提案 → サマリー + 改善提案（15行以内） + - REJECT → 問題点を表形式で（30行以内） + allowed_tools: + - Read + - Glob + - Grep + - Write + - WebSearch + - WebFetch + instruction_template: | + **アーキテクチャと設計**のレビューに集中してください。AI特有の問題はレビューしないでください（前のステップで完了済み）。 + + 変更をレビューしてフィードバックを提供してください。 + + **注意:** simpleワークフローではIMPROVE判定は使用しません。 + 軽微な改善提案がある場合は APPROVE + コメントとしてください。 + rules: + - condition: "問題なし" + next: supervise + - condition: "構造的な修正必要" + next: plan - name: supervise + edit: false agent: ~/.takt/agents/default/supervisor.md + report: + - Validation: 05-supervisor-validation.md + - Summary: summary.md allowed_tools: - Read - Glob @@ -458,65 +250,7 @@ steps: - Bash - WebSearch - WebFetch - status_rules_prompt: | - ## 判定基準 - - | 状況 | 判定 | - |------|------| - | 要求が満たされていない | REJECT | - | テストが失敗する | REJECT | - | ビルドが通らない | REJECT | - | その場しのぎが残っている | REJECT | - | すべて問題なし | APPROVE | - - **原則**: 疑わしきは REJECT。曖昧な承認はしない。 - - ## 出力フォーマット - - | 状況 | タグ | - |------|------| - | 最終承認 | `[SUPERVISOR:APPROVE]` | - | 差し戻し | `[SUPERVISOR:REJECT]` | - - ### 出力例 - - **APPROVE の場合:** - ``` - [SUPERVISOR:APPROVE] - - 検証結果: - - テスト: ✅ 全件パス - - ビルド: ✅ 成功 - - 要求充足: ✅ - ``` - - **REJECT の場合:** - ``` - [SUPERVISOR:REJECT] - - 問題点: - 1. テストが失敗: `npm test` で 2 件失敗 - 2. 元の要求を満たしていない: ログイン機能が未実装 - ``` instruction_template: | - ## Workflow Context - - Iteration: {iteration}/{max_iterations}（ワークフロー全体） - - Step Iteration: {step_iteration}（このステップの実行回数） - - Step: supervise (final verification) - - Report Directory: .takt/reports/{report_dir}/ - - Report Files: - - Validation: .takt/reports/{report_dir}/05-supervisor-validation.md - - Summary: .takt/reports/{report_dir}/summary.md - - ## Original User Request - {task} - - ## Git Diff - ```diff - {git_diff} - ``` - - ## Instructions テスト実行、ビルド確認、最終承認を行ってください。 **ワークフロー全体の確認:** @@ -527,10 +261,6 @@ steps: **レポートの確認:** Report Directory内の全レポートを読み、 未対応の改善提案がないか確認してください。 - **レポート出力:** 上記の `Report Files` に出力してください。 - - ファイルが存在しない場合: 新規作成 - - ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記 - **Validationレポートフォーマット:** ```markdown # 最終検証結果 @@ -573,8 +303,8 @@ steps: ## レビュー結果 | レビュー | 結果 | |---------|------| - | Architect | ✅ APPROVE | | AI Review | ✅ APPROVE | + | Architect | ✅ APPROVE | | Supervisor | ✅ APPROVE | ## 確認コマンド @@ -583,8 +313,8 @@ steps: npm run build ``` ``` - transitions: - - condition: approved - next_step: COMPLETE - - condition: rejected - next_step: plan + rules: + - condition: "すべて問題なし" + next: COMPLETE + - condition: "要求未達成、テスト失敗、ビルドエラー" + next: plan diff --git a/src/__tests__/client.test.ts b/src/__tests__/client.test.ts index 6902ac9..55611e1 100644 --- a/src/__tests__/client.test.ts +++ b/src/__tests__/client.test.ts @@ -4,43 +4,10 @@ import { describe, it, expect } from 'vitest'; import { - detectStatus, + detectRuleIndex, isRegexSafe, - getBuiltinStatusPatterns, } from '../claude/client.js'; -describe('detectStatus', () => { - it('should detect done status', () => { - const content = 'Task completed successfully.\n[CODER:DONE]'; - const patterns = { done: '\\[CODER:DONE\\]' }; - expect(detectStatus(content, patterns)).toBe('done'); - }); - - it('should detect approved status', () => { - const content = 'Code looks good.\n[ARCHITECT:APPROVE]'; - const patterns = { approved: '\\[ARCHITECT:APPROVE\\]' }; - expect(detectStatus(content, patterns)).toBe('approved'); - }); - - it('should return in_progress when no pattern matches', () => { - const content = 'Working on it...'; - const patterns = { done: '\\[DONE\\]' }; - expect(detectStatus(content, patterns)).toBe('in_progress'); - }); - - it('should be case insensitive', () => { - const content = '[coder:done]'; - const patterns = { done: '\\[CODER:DONE\\]' }; - expect(detectStatus(content, patterns)).toBe('done'); - }); - - it('should handle invalid regex gracefully', () => { - const content = 'test'; - const patterns = { done: '[invalid(' }; - expect(detectStatus(content, patterns)).toBe('in_progress'); - }); -}); - describe('isRegexSafe', () => { it('should accept simple patterns', () => { expect(isRegexSafe('\\[DONE\\]')).toBe(true); @@ -60,32 +27,43 @@ describe('isRegexSafe', () => { }); }); -describe('getBuiltinStatusPatterns', () => { - it('should return patterns for coder', () => { - const patterns = getBuiltinStatusPatterns('coder'); - expect(patterns.done).toBeDefined(); - expect(patterns.blocked).toBeDefined(); +describe('detectRuleIndex', () => { + it('should detect [PLAN:1] as index 0', () => { + expect(detectRuleIndex('Analysis complete.\n[PLAN:1]', 'plan')).toBe(0); }); - it('should return patterns for architect', () => { - const patterns = getBuiltinStatusPatterns('architect'); - expect(patterns.approved).toBeDefined(); - expect(patterns.rejected).toBeDefined(); + it('should detect [PLAN:2] as index 1', () => { + expect(detectRuleIndex('Question answered.\n[PLAN:2]', 'plan')).toBe(1); }); - it('should return generic patterns for unknown agent', () => { - // With generic patterns, any agent gets the standard patterns - const patterns = getBuiltinStatusPatterns('unknown'); - expect(patterns.approved).toBeDefined(); - expect(patterns.rejected).toBeDefined(); - expect(patterns.done).toBeDefined(); - expect(patterns.blocked).toBeDefined(); + it('should detect [PLAN:3] as index 2', () => { + expect(detectRuleIndex('[PLAN:3]\n\nBlocked.', 'plan')).toBe(2); }); - it('should detect status using generic patterns for any agent', () => { - // Generic patterns work for any [ROLE:COMMAND] format - const patterns = getBuiltinStatusPatterns('my-custom-agent'); - const content = '[MY_CUSTOM_AGENT:APPROVE]'; - expect(detectStatus(content, patterns)).toBe('approved'); + it('should be case insensitive', () => { + expect(detectRuleIndex('[plan:1]', 'plan')).toBe(0); + expect(detectRuleIndex('[Plan:2]', 'plan')).toBe(1); + }); + + it('should match step name case-insensitively', () => { + expect(detectRuleIndex('[IMPLEMENT:1]', 'implement')).toBe(0); + expect(detectRuleIndex('[REVIEW:2]', 'review')).toBe(1); + }); + + it('should return -1 when no match', () => { + expect(detectRuleIndex('No tags here.', 'plan')).toBe(-1); + }); + + it('should return -1 for [PLAN:0] (invalid)', () => { + expect(detectRuleIndex('[PLAN:0]', 'plan')).toBe(-1); + }); + + it('should return -1 for wrong step name', () => { + expect(detectRuleIndex('[REVIEW:1]', 'plan')).toBe(-1); + }); + + it('should handle step names with hyphens', () => { + expect(detectRuleIndex('[AI_REVIEW:1]', 'ai_review')).toBe(0); + expect(detectRuleIndex('[SECURITY_FIX:2]', 'security_fix')).toBe(1); }); }); diff --git a/src/__tests__/config.test.ts b/src/__tests__/config.test.ts index 558c9a3..a6f306e 100644 --- a/src/__tests__/config.test.ts +++ b/src/__tests__/config.test.ts @@ -65,9 +65,9 @@ steps: - name: step1 agent: coder instruction: "{task}" - transitions: - - condition: done - next_step: COMPLETE + rules: + - condition: Task completed + next: COMPLETE `; writeFileSync(join(workflowsDir, 'test.yaml'), sampleWorkflow); diff --git a/src/__tests__/engine-report.test.ts b/src/__tests__/engine-report.test.ts new file mode 100644 index 0000000..2bd0a09 --- /dev/null +++ b/src/__tests__/engine-report.test.ts @@ -0,0 +1,175 @@ +/** + * Tests for engine report event emission (step:report) + */ + +import { describe, it, expect, beforeEach, afterEach, vi } from 'vitest'; +import { mkdirSync, rmSync, writeFileSync } from 'node:fs'; +import { join } from 'node:path'; +import { tmpdir } from 'node:os'; +import { EventEmitter } from 'node:events'; +import { existsSync } from 'node:fs'; +import { isReportObjectConfig } from '../workflow/instruction-builder.js'; +import type { WorkflowStep, ReportObjectConfig, ReportConfig } from '../models/types.js'; + +/** + * Extracted emitStepReports logic for unit testing. + * Mirrors engine.ts emitStepReports + emitIfReportExists. + */ +function emitStepReports( + emitter: EventEmitter, + step: WorkflowStep, + reportDir: string, + projectCwd: string, +): void { + if (!step.report || !reportDir) return; + const baseDir = join(projectCwd, '.takt', 'reports', reportDir); + + if (typeof step.report === 'string') { + emitIfReportExists(emitter, step, baseDir, step.report); + } else if (isReportObjectConfig(step.report)) { + emitIfReportExists(emitter, step, baseDir, step.report.name); + } else { + for (const rc of step.report) { + emitIfReportExists(emitter, step, baseDir, rc.path); + } + } +} + +function emitIfReportExists( + emitter: EventEmitter, + step: WorkflowStep, + baseDir: string, + fileName: string, +): void { + const filePath = join(baseDir, fileName); + if (existsSync(filePath)) { + emitter.emit('step:report', step, filePath, fileName); + } +} + +/** Create a minimal WorkflowStep for testing */ +function createStep(overrides: Partial = {}): WorkflowStep { + return { + name: 'test-step', + agent: 'coder', + agentDisplayName: 'Coder', + instructionTemplate: '', + passPreviousResponse: false, + ...overrides, + }; +} + +describe('emitStepReports', () => { + let tmpDir: string; + let reportBaseDir: string; + const reportDirName = 'test-report-dir'; + + beforeEach(() => { + tmpDir = join(tmpdir(), `takt-report-test-${Date.now()}`); + reportBaseDir = join(tmpDir, '.takt', 'reports', reportDirName); + mkdirSync(reportBaseDir, { recursive: true }); + }); + + afterEach(() => { + rmSync(tmpDir, { recursive: true, force: true }); + }); + + it('should emit step:report when string report file exists', () => { + // Given: a step with string report and the file exists + const step = createStep({ report: 'plan.md' }); + writeFileSync(join(reportBaseDir, 'plan.md'), '# Plan', 'utf-8'); + const emitter = new EventEmitter(); + const handler = vi.fn(); + emitter.on('step:report', handler); + + // When + emitStepReports(emitter, step, reportDirName, tmpDir); + + // Then + expect(handler).toHaveBeenCalledOnce(); + expect(handler).toHaveBeenCalledWith(step, join(reportBaseDir, 'plan.md'), 'plan.md'); + }); + + it('should not emit when string report file does not exist', () => { + // Given: a step with string report but file doesn't exist + const step = createStep({ report: 'missing.md' }); + const emitter = new EventEmitter(); + const handler = vi.fn(); + emitter.on('step:report', handler); + + // When + emitStepReports(emitter, step, reportDirName, tmpDir); + + // Then + expect(handler).not.toHaveBeenCalled(); + }); + + it('should emit step:report when ReportObjectConfig report file exists', () => { + // Given: a step with ReportObjectConfig and the file exists + const report: ReportObjectConfig = { name: '03-review.md', format: '# Review' }; + const step = createStep({ report }); + writeFileSync(join(reportBaseDir, '03-review.md'), '# Review\nOK', 'utf-8'); + const emitter = new EventEmitter(); + const handler = vi.fn(); + emitter.on('step:report', handler); + + // When + emitStepReports(emitter, step, reportDirName, tmpDir); + + // Then + expect(handler).toHaveBeenCalledOnce(); + expect(handler).toHaveBeenCalledWith(step, join(reportBaseDir, '03-review.md'), '03-review.md'); + }); + + it('should emit for each existing file in ReportConfig[] array', () => { + // Given: a step with array report, two files exist, one missing + const report: ReportConfig[] = [ + { label: 'Scope', path: '01-scope.md' }, + { label: 'Decisions', path: '02-decisions.md' }, + { label: 'Missing', path: '03-missing.md' }, + ]; + const step = createStep({ report }); + writeFileSync(join(reportBaseDir, '01-scope.md'), '# Scope', 'utf-8'); + writeFileSync(join(reportBaseDir, '02-decisions.md'), '# Decisions', 'utf-8'); + const emitter = new EventEmitter(); + const handler = vi.fn(); + emitter.on('step:report', handler); + + // When + emitStepReports(emitter, step, reportDirName, tmpDir); + + // Then: emitted for scope and decisions, not for missing + expect(handler).toHaveBeenCalledTimes(2); + expect(handler).toHaveBeenCalledWith(step, join(reportBaseDir, '01-scope.md'), '01-scope.md'); + expect(handler).toHaveBeenCalledWith(step, join(reportBaseDir, '02-decisions.md'), '02-decisions.md'); + }); + + it('should not emit when step has no report', () => { + // Given: a step without report + const step = createStep({ report: undefined }); + const emitter = new EventEmitter(); + const handler = vi.fn(); + emitter.on('step:report', handler); + + // When + emitStepReports(emitter, step, reportDirName, tmpDir); + + // Then + expect(handler).not.toHaveBeenCalled(); + }); + + it('should not emit when reportDir is empty', () => { + // Given: a step with report but empty reportDir + const step = createStep({ report: 'plan.md' }); + writeFileSync(join(reportBaseDir, 'plan.md'), '# Plan', 'utf-8'); + const emitter = new EventEmitter(); + const handler = vi.fn(); + emitter.on('step:report', handler); + + // When: empty reportDir + emitStepReports(emitter, step, '', tmpDir); + + // Then + expect(handler).not.toHaveBeenCalled(); + }); +}); diff --git a/src/__tests__/instructionBuilder.test.ts b/src/__tests__/instructionBuilder.test.ts index b5c906d..77e9eaa 100644 --- a/src/__tests__/instructionBuilder.test.ts +++ b/src/__tests__/instructionBuilder.test.ts @@ -8,9 +8,12 @@ import { buildExecutionMetadata, renderExecutionMetadata, renderStatusRulesHeader, + generateStatusRulesFromRules, + isReportObjectConfig, type InstructionContext, } from '../workflow/instruction-builder.js'; -import type { WorkflowStep } from '../models/types.js'; +import type { WorkflowStep, WorkflowRule } from '../models/types.js'; + function createMinimalStep(template: string): WorkflowStep { return { @@ -18,7 +21,6 @@ function createMinimalStep(template: string): WorkflowStep { agent: 'test-agent', agentDisplayName: 'Test Agent', instructionTemplate: template, - transitions: [], passPreviousResponse: false, }; } @@ -74,6 +76,34 @@ describe('instruction-builder', () => { expect(metadataIndex).toBeLessThan(bodyIndex); }); + + it('should include edit enabled prompt when step.edit is true', () => { + const step = { ...createMinimalStep('Implement feature'), edit: true as const }; + const context = createMinimalContext({ cwd: '/project' }); + + const result = buildInstruction(step, context); + + expect(result).toContain('Editing is ENABLED'); + }); + + it('should include edit disabled prompt when step.edit is false', () => { + const step = { ...createMinimalStep('Review code'), edit: false as const }; + const context = createMinimalContext({ cwd: '/project' }); + + const result = buildInstruction(step, context); + + expect(result).toContain('Editing is DISABLED'); + }); + + it('should not include edit prompt when step.edit is undefined', () => { + const step = createMinimalStep('Do some work'); + const context = createMinimalContext({ cwd: '/project' }); + + const result = buildInstruction(step, context); + + expect(result).not.toContain('Editing is ENABLED'); + expect(result).not.toContain('Editing is DISABLED'); + }); }); describe('report_dir replacement', () => { @@ -175,6 +205,20 @@ describe('instruction-builder', () => { expect(metadata.language).toBe('ja'); }); + + it('should propagate edit field when provided', () => { + const context = createMinimalContext({ cwd: '/project' }); + + expect(buildExecutionMetadata(context, true).edit).toBe(true); + expect(buildExecutionMetadata(context, false).edit).toBe(false); + }); + + it('should leave edit undefined when not provided', () => { + const context = createMinimalContext({ cwd: '/project' }); + const metadata = buildExecutionMetadata(context); + + expect(metadata.edit).toBeUndefined(); + }); }); describe('renderExecutionMetadata', () => { @@ -211,6 +255,43 @@ describe('instruction-builder', () => { expect(enRendered).toContain('Note:'); expect(jaRendered).not.toContain('Note:'); }); + + it('should include edit enabled prompt when edit is true (en)', () => { + const rendered = renderExecutionMetadata({ workingDirectory: '/project', language: 'en', edit: true }); + + expect(rendered).toContain('Editing is ENABLED'); + expect(rendered).not.toContain('Editing is DISABLED'); + }); + + it('should include edit disabled prompt when edit is false (en)', () => { + const rendered = renderExecutionMetadata({ workingDirectory: '/project', language: 'en', edit: false }); + + expect(rendered).toContain('Editing is DISABLED'); + expect(rendered).not.toContain('Editing is ENABLED'); + }); + + it('should not include edit prompt when edit is undefined', () => { + const rendered = renderExecutionMetadata({ workingDirectory: '/project', language: 'en' }); + + expect(rendered).not.toContain('Editing is ENABLED'); + expect(rendered).not.toContain('Editing is DISABLED'); + expect(rendered).not.toContain('編集が許可'); + expect(rendered).not.toContain('編集が禁止'); + }); + + it('should render edit enabled prompt in Japanese when language is ja', () => { + const rendered = renderExecutionMetadata({ workingDirectory: '/project', language: 'ja', edit: true }); + + expect(rendered).toContain('編集が許可されています'); + expect(rendered).not.toContain('編集が禁止'); + }); + + it('should render edit disabled prompt in Japanese when language is ja', () => { + const rendered = renderExecutionMetadata({ workingDirectory: '/project', language: 'ja', edit: false }); + + expect(rendered).toContain('編集が禁止されています'); + expect(rendered).not.toContain('編集が許可'); + }); }); describe('renderStatusRulesHeader', () => { @@ -237,41 +318,566 @@ describe('instruction-builder', () => { }); }); - describe('status_rules_prompt with header', () => { - it('should prepend status header to status_rules_prompt in Japanese', () => { - const step = createMinimalStep('Do work'); - step.statusRulesPrompt = '## 出力フォーマット\n| 状況 | タグ |'; - const context = createMinimalContext({ language: 'ja' }); + describe('generateStatusRulesFromRules', () => { + const rules: WorkflowRule[] = [ + { condition: '要件が明確で実装可能', next: 'implement' }, + { condition: 'ユーザーが質問をしている', next: 'COMPLETE' }, + { condition: '要件が不明確、情報不足', next: 'ABORT', appendix: '確認事項:\n- {質問1}\n- {質問2}' }, + ]; - const result = buildInstruction(step, context); + it('should generate criteria table with numbered tags (ja)', () => { + const result = generateStatusRulesFromRules('plan', rules, 'ja'); - expect(result).toContain('# ⚠️ 必須: ステータス出力ルール ⚠️'); - expect(result).toContain('## 出力フォーマット'); - // Header should come before the step-specific content - const headerIndex = result.indexOf('⚠️ 必須'); - const formatIndex = result.indexOf('## 出力フォーマット'); - expect(headerIndex).toBeLessThan(formatIndex); + expect(result).toContain('## 判定基準'); + expect(result).toContain('| 1 | 要件が明確で実装可能 | `[PLAN:1]` |'); + expect(result).toContain('| 2 | ユーザーが質問をしている | `[PLAN:2]` |'); + expect(result).toContain('| 3 | 要件が不明確、情報不足 | `[PLAN:3]` |'); }); - it('should prepend status header to status_rules_prompt in English', () => { + it('should generate criteria table with numbered tags (en)', () => { + const enRules: WorkflowRule[] = [ + { condition: 'Requirements are clear', next: 'implement' }, + { condition: 'User is asking a question', next: 'COMPLETE' }, + ]; + const result = generateStatusRulesFromRules('plan', enRules, 'en'); + + expect(result).toContain('## Decision Criteria'); + expect(result).toContain('| 1 | Requirements are clear | `[PLAN:1]` |'); + expect(result).toContain('| 2 | User is asking a question | `[PLAN:2]` |'); + }); + + it('should generate output format section with condition labels', () => { + const result = generateStatusRulesFromRules('plan', rules, 'ja'); + + expect(result).toContain('## 出力フォーマット'); + expect(result).toContain('`[PLAN:1]` — 要件が明確で実装可能'); + expect(result).toContain('`[PLAN:2]` — ユーザーが質問をしている'); + expect(result).toContain('`[PLAN:3]` — 要件が不明確、情報不足'); + }); + + it('should generate appendix template section when rules have appendix', () => { + const result = generateStatusRulesFromRules('plan', rules, 'ja'); + + expect(result).toContain('### 追加出力テンプレート'); + expect(result).toContain('`[PLAN:3]`'); + expect(result).toContain('確認事項:'); + expect(result).toContain('- {質問1}'); + }); + + it('should not generate appendix section when no rules have appendix', () => { + const noAppendixRules: WorkflowRule[] = [ + { condition: 'Done', next: 'review' }, + { condition: 'Blocked', next: 'plan' }, + ]; + const result = generateStatusRulesFromRules('implement', noAppendixRules, 'en'); + + expect(result).not.toContain('Appendix Template'); + }); + + it('should uppercase step name in tags', () => { + const result = generateStatusRulesFromRules('ai_review', [ + { condition: 'No issues', next: 'supervise' }, + ], 'en'); + + expect(result).toContain('`[AI_REVIEW:1]`'); + }); + }); + + describe('buildInstruction with rules', () => { + it('should auto-generate status rules from rules', () => { const step = createMinimalStep('Do work'); - step.statusRulesPrompt = '## Output Format\n| Situation | Tag |'; + step.name = 'plan'; + step.rules = [ + { condition: 'Clear requirements', next: 'implement' }, + { condition: 'Unclear', next: 'ABORT' }, + ]; const context = createMinimalContext({ language: 'en' }); const result = buildInstruction(step, context); - expect(result).toContain('# ⚠️ Required: Status Output Rules ⚠️'); - expect(result).toContain('## Output Format'); + // Should contain status header + expect(result).toContain('⚠️ Required: Status Output Rules ⚠️'); + // Should contain auto-generated criteria table + expect(result).toContain('## Decision Criteria'); + expect(result).toContain('`[PLAN:1]`'); + expect(result).toContain('`[PLAN:2]`'); }); - it('should not add header if statusRulesPrompt is not set', () => { + it('should not add status rules when rules do not exist', () => { const step = createMinimalStep('Do work'); const context = createMinimalContext({ language: 'en' }); const result = buildInstruction(step, context); expect(result).not.toContain('⚠️ Required'); - expect(result).not.toContain('⚠️ 必須'); + expect(result).not.toContain('Decision Criteria'); + }); + + it('should not auto-generate when rules array is empty', () => { + const step = createMinimalStep('Do work'); + step.rules = []; + const context = createMinimalContext({ language: 'en' }); + + const result = buildInstruction(step, context); + + expect(result).not.toContain('⚠️ Required'); + expect(result).not.toContain('Decision Criteria'); + }); + }); + + describe('auto-injected Workflow Context section', () => { + it('should include iteration, step iteration, and step name', () => { + const step = createMinimalStep('Do work'); + step.name = 'implement'; + const context = createMinimalContext({ + iteration: 3, + maxIterations: 20, + stepIteration: 2, + language: 'en', + }); + + const result = buildInstruction(step, context); + + expect(result).toContain('## Workflow Context'); + expect(result).toContain('- Iteration: 3/20'); + expect(result).toContain('- Step Iteration: 2'); + expect(result).toContain('- Step: implement'); + }); + + it('should include single report file when report is a string', () => { + const step = createMinimalStep('Do work'); + step.name = 'plan'; + step.report = '00-plan.md'; + const context = createMinimalContext({ + reportDir: '20260129-test', + language: 'en', + }); + + const result = buildInstruction(step, context); + + expect(result).toContain('- Report Directory: 20260129-test/'); + expect(result).toContain('- Report File: 20260129-test/00-plan.md'); + expect(result).not.toContain('Report Files:'); + }); + + it('should include multiple report files when report is ReportConfig[]', () => { + const step = createMinimalStep('Do work'); + step.name = 'implement'; + step.report = [ + { label: 'Scope', path: '01-scope.md' }, + { label: 'Decisions', path: '02-decisions.md' }, + ]; + const context = createMinimalContext({ + reportDir: '20260129-test', + language: 'en', + }); + + const result = buildInstruction(step, context); + + expect(result).toContain('- Report Directory: 20260129-test/'); + expect(result).toContain('- Report Files:'); + expect(result).toContain(' - Scope: 20260129-test/01-scope.md'); + expect(result).toContain(' - Decisions: 20260129-test/02-decisions.md'); + expect(result).not.toContain('Report File:'); + }); + + it('should include report file when report is ReportObjectConfig', () => { + const step = createMinimalStep('Do work'); + step.name = 'plan'; + step.report = { name: '00-plan.md' }; + const context = createMinimalContext({ + reportDir: '20260129-test', + language: 'en', + }); + + const result = buildInstruction(step, context); + + expect(result).toContain('- Report Directory: 20260129-test/'); + expect(result).toContain('- Report File: 20260129-test/00-plan.md'); + expect(result).not.toContain('Report Files:'); + }); + + it('should NOT include report info when reportDir is undefined', () => { + const step = createMinimalStep('Do work'); + step.report = '00-plan.md'; + const context = createMinimalContext({ language: 'en' }); + + const result = buildInstruction(step, context); + + expect(result).toContain('## Workflow Context'); + expect(result).not.toContain('Report Directory'); + expect(result).not.toContain('Report File'); + }); + + it('should NOT include report info when step has no report', () => { + const step = createMinimalStep('Do work'); + const context = createMinimalContext({ + reportDir: '20260129-test', + language: 'en', + }); + + const result = buildInstruction(step, context); + + expect(result).toContain('## Workflow Context'); + expect(result).not.toContain('Report Directory'); + expect(result).not.toContain('Report File'); + }); + + it('should render Japanese step iteration suffix', () => { + const step = createMinimalStep('Do work'); + const context = createMinimalContext({ + stepIteration: 3, + language: 'ja', + }); + + const result = buildInstruction(step, context); + + expect(result).toContain('- Step Iteration: 3（このステップの実行回数）'); + }); + + it('should NOT include .takt/reports/ prefix in report paths', () => { + const step = createMinimalStep('Do work'); + step.report = '00-plan.md'; + const context = createMinimalContext({ + reportDir: '20260129-test', + language: 'en', + }); + + const result = buildInstruction(step, context); + + expect(result).not.toContain('.takt/reports/'); + }); + }); + + describe('ReportObjectConfig order/format injection', () => { + it('should inject order before instruction_template', () => { + const step = createMinimalStep('Do work'); + step.report = { + name: '00-plan.md', + order: '**Output:** Write to {report:00-plan.md}', + }; + const context = createMinimalContext({ + reportDir: '20260129-test', + language: 'en', + }); + + const result = buildInstruction(step, context); + + const orderIdx = result.indexOf('**Output:** Write to 20260129-test/00-plan.md'); + const instructionsIdx = result.indexOf('## Instructions'); + expect(orderIdx).toBeGreaterThan(-1); + expect(instructionsIdx).toBeGreaterThan(orderIdx); + }); + + it('should inject format after instruction_template', () => { + const step = createMinimalStep('Do work'); + step.report = { + name: '00-plan.md', + format: '**Format:**\n```markdown\n# Plan\n```', + }; + const context = createMinimalContext({ + reportDir: '20260129-test', + language: 'en', + }); + + const result = buildInstruction(step, context); + + const instructionsIdx = result.indexOf('## Instructions'); + const formatIdx = result.indexOf('**Format:**'); + expect(formatIdx).toBeGreaterThan(instructionsIdx); + }); + + it('should inject both order before and format after instruction_template', () => { + const step = createMinimalStep('Do work'); + step.report = { + name: '00-plan.md', + order: '**Output:** Write to {report:00-plan.md}', + format: '**Format:**\n```markdown\n# Plan\n```', + }; + const context = createMinimalContext({ + reportDir: '20260129-test', + language: 'en', + }); + + const result = buildInstruction(step, context); + + const orderIdx = result.indexOf('**Output:** Write to 20260129-test/00-plan.md'); + const instructionsIdx = result.indexOf('## Instructions'); + const formatIdx = result.indexOf('**Format:**'); + expect(orderIdx).toBeGreaterThan(-1); + expect(instructionsIdx).toBeGreaterThan(orderIdx); + expect(formatIdx).toBeGreaterThan(instructionsIdx); + }); + + it('should replace {report:filename} in order text', () => { + const step = createMinimalStep('Do work'); + step.report = { + name: '00-plan.md', + order: 'Output to {report:00-plan.md} file.', + }; + const context = createMinimalContext({ + reportDir: '20260129-test', + language: 'en', + }); + + const result = buildInstruction(step, context); + + expect(result).toContain('Output to 20260129-test/00-plan.md file.'); + expect(result).not.toContain('{report:00-plan.md}'); + }); + + it('should auto-inject report output instruction when report is a simple string', () => { + const step = createMinimalStep('Do work'); + step.report = '00-plan.md'; + const context = createMinimalContext({ + reportDir: '20260129-test', + language: 'en', + }); + + const result = buildInstruction(step, context); + + // Auto-generated report output instruction should be injected before ## Instructions + expect(result).toContain('**Report output:** Output to the `Report File` specified above.'); + expect(result).toContain('- If file does not exist: Create new file'); + const reportIdx = result.indexOf('**Report output:**'); + const instructionsIdx = result.indexOf('## Instructions'); + expect(reportIdx).toBeGreaterThan(-1); + expect(instructionsIdx).toBeGreaterThan(reportIdx); + }); + + it('should auto-inject report output instruction when report is ReportConfig[]', () => { + const step = createMinimalStep('Do work'); + step.report = [ + { label: 'Scope', path: '01-scope.md' }, + ]; + const context = createMinimalContext({ + reportDir: '20260129-test', + language: 'en', + }); + + const result = buildInstruction(step, context); + + // Auto-generated multi-file report output instruction + expect(result).toContain('**Report output:** Output to the `Report Files` specified above.'); + expect(result).toContain('- If file does not exist: Create new file'); + }); + + it('should replace {report:filename} in instruction_template too', () => { + const step = createMinimalStep('Write to {report:00-plan.md}'); + const context = createMinimalContext({ + reportDir: '20260129-test', + language: 'en', + }); + + const result = buildInstruction(step, context); + + expect(result).toContain('Write to 20260129-test/00-plan.md'); + expect(result).not.toContain('{report:00-plan.md}'); + }); + + it('should replace {step_iteration} in order/format text', () => { + const step = createMinimalStep('Do work'); + step.report = { + name: '00-plan.md', + order: 'Append ## Iteration {step_iteration} section', + }; + const context = createMinimalContext({ + reportDir: '20260129-test', + stepIteration: 3, + language: 'en', + }); + + const result = buildInstruction(step, context); + + expect(result).toContain('Append ## Iteration 3 section'); + }); + + it('should auto-inject Japanese report output instruction for ja language', () => { + const step = createMinimalStep('作業する'); + step.report = { name: '00-plan.md' }; + const context = createMinimalContext({ + reportDir: '20260129-test', + language: 'ja', + }); + + const result = buildInstruction(step, context); + + expect(result).toContain('**レポート出力:** `Report File` に出力してください。'); + expect(result).toContain('- ファイルが存在しない場合: 新規作成'); + expect(result).toContain('- ファイルが存在する場合: `## Iteration 1` セクションを追記'); + }); + + it('should auto-inject Japanese multi-file report output instruction', () => { + const step = createMinimalStep('作業する'); + step.report = [{ label: 'Scope', path: '01-scope.md' }]; + const context = createMinimalContext({ + reportDir: '20260129-test', + language: 'ja', + }); + + const result = buildInstruction(step, context); + + expect(result).toContain('**レポート出力:** Report Files に出力してください。'); + }); + + it('should replace {step_iteration} in auto-generated report output instruction', () => { + const step = createMinimalStep('Do work'); + step.report = '00-plan.md'; + const context = createMinimalContext({ + reportDir: '20260129-test', + stepIteration: 5, + language: 'en', + }); + + const result = buildInstruction(step, context); + + expect(result).toContain('Append with `## Iteration 5` section'); + }); + + it('should prefer explicit order over auto-generated report instruction', () => { + const step = createMinimalStep('Do work'); + step.report = { + name: '00-plan.md', + order: 'Custom order instruction', + }; + const context = createMinimalContext({ + reportDir: '20260129-test', + language: 'en', + }); + + const result = buildInstruction(step, context); + + expect(result).toContain('Custom order instruction'); + expect(result).not.toContain('**Report output:**'); + }); + + it('should auto-inject report output for ReportObjectConfig without order', () => { + const step = createMinimalStep('Do work'); + step.report = { name: '00-plan.md', format: '# Plan' }; + const context = createMinimalContext({ + reportDir: '20260129-test', + language: 'en', + }); + + const result = buildInstruction(step, context); + + expect(result).toContain('**Report output:** Output to the `Report File` specified above.'); + }); + + it('should NOT inject report output when no reportDir', () => { + const step = createMinimalStep('Do work'); + step.report = '00-plan.md'; + const context = createMinimalContext({ + language: 'en', + }); + + const result = buildInstruction(step, context); + + expect(result).not.toContain('**Report output:**'); + }); + }); + + describe('auto-injected User Request and Additional User Inputs sections', () => { + it('should include User Request section with task', () => { + const step = createMinimalStep('Do work'); + const context = createMinimalContext({ task: 'Build the feature', language: 'en' }); + + const result = buildInstruction(step, context); + + expect(result).toContain('## User Request\n'); + }); + + it('should include Additional User Inputs section', () => { + const step = createMinimalStep('Do work'); + const context = createMinimalContext({ + userInputs: ['input1', 'input2'], + language: 'en', + }); + + const result = buildInstruction(step, context); + + expect(result).toContain('## Additional User Inputs\n'); + }); + + it('should include Previous Response when passPreviousResponse is true and output exists', () => { + const step = createMinimalStep('Do work'); + step.passPreviousResponse = true; + const context = createMinimalContext({ + previousOutput: { content: 'Previous result', tag: '[TEST:1]' }, + language: 'en', + }); + + const result = buildInstruction(step, context); + + expect(result).toContain('## Previous Response\n'); + }); + + it('should NOT include Previous Response when passPreviousResponse is false', () => { + const step = createMinimalStep('Do work'); + step.passPreviousResponse = false; + const context = createMinimalContext({ + previousOutput: { content: 'Previous result', tag: '[TEST:1]' }, + language: 'en', + }); + + const result = buildInstruction(step, context); + + expect(result).not.toContain('## Previous Response'); + }); + + it('should include Instructions header before template content', () => { + const step = createMinimalStep('My specific instructions here'); + const context = createMinimalContext({ language: 'en' }); + + const result = buildInstruction(step, context); + + const instructionsIdx = result.indexOf('## Instructions'); + const contentIdx = result.indexOf('My specific instructions here'); + expect(instructionsIdx).toBeGreaterThan(-1); + expect(contentIdx).toBeGreaterThan(instructionsIdx); + }); + + it('should skip auto-injected User Request when template contains {task}', () => { + const step = createMinimalStep('Process this: {task}'); + const context = createMinimalContext({ task: 'My task', language: 'en' }); + + const result = buildInstruction(step, context); + + // Auto-injected section should NOT appear + expect(result).not.toContain('## User Request'); + // But template placeholder should be replaced + expect(result).toContain('Process this: My task'); + }); + + it('should skip auto-injected Previous Response when template contains {previous_response}', () => { + const step = createMinimalStep('## Feedback\n{previous_response}\n\nFix the issues.'); + step.passPreviousResponse = true; + const context = createMinimalContext({ + previousOutput: { content: 'Review feedback here', tag: '[TEST:1]' }, + language: 'en', + }); + + const result = buildInstruction(step, context); + + // Auto-injected section should NOT appear + expect(result).not.toContain('## Previous Response\n'); + // But template placeholder should be replaced with content + expect(result).toContain('## Feedback\nReview feedback here'); + }); + + it('should skip auto-injected Additional User Inputs when template contains {user_inputs}', () => { + const step = createMinimalStep('Inputs: {user_inputs}'); + const context = createMinimalContext({ + userInputs: ['extra info'], + language: 'en', + }); + + const result = buildInstruction(step, context); + + // Auto-injected section should NOT appear + expect(result).not.toContain('## Additional User Inputs'); + // But template placeholder should be replaced + expect(result).toContain('Inputs: extra info'); }); }); @@ -303,4 +909,22 @@ describe('instruction-builder', () => { expect(result).toContain('Run #2'); }); }); + + describe('isReportObjectConfig', () => { + it('should return true for ReportObjectConfig', () => { + expect(isReportObjectConfig({ name: '00-plan.md' })).toBe(true); + }); + + it('should return true for ReportObjectConfig with order/format', () => { + expect(isReportObjectConfig({ name: '00-plan.md', order: 'output to...', format: '# Plan' })).toBe(true); + }); + + it('should return false for string', () => { + expect(isReportObjectConfig('00-plan.md')).toBe(false); + }); + + it('should return false for ReportConfig[] (array)', () => { + expect(isReportObjectConfig([{ label: 'Scope', path: '01-scope.md' }])).toBe(false); + }); + }); }); diff --git a/src/__tests__/models.test.ts b/src/__tests__/models.test.ts index f470c83..dcd9ef5 100644 --- a/src/__tests__/models.test.ts +++ b/src/__tests__/models.test.ts @@ -6,12 +6,10 @@ import { describe, it, expect } from 'vitest'; import { AgentTypeSchema, StatusSchema, - TransitionConditionSchema, PermissionModeSchema, WorkflowConfigRawSchema, CustomAgentConfigSchema, GlobalConfigSchema, - GENERIC_STATUS_PATTERNS, } from '../models/schemas.js'; describe('AgentTypeSchema', () => { @@ -43,21 +41,6 @@ describe('StatusSchema', () => { }); }); -describe('TransitionConditionSchema', () => { - it('should accept valid conditions', () => { - expect(TransitionConditionSchema.parse('done')).toBe('done'); - expect(TransitionConditionSchema.parse('approved')).toBe('approved'); - expect(TransitionConditionSchema.parse('rejected')).toBe('rejected'); - expect(TransitionConditionSchema.parse('always')).toBe('always'); - expect(TransitionConditionSchema.parse('answer')).toBe('answer'); - }); - - it('should reject invalid conditions', () => { - expect(() => TransitionConditionSchema.parse('conditional')).toThrow(); - expect(() => TransitionConditionSchema.parse('fixed')).toThrow(); - }); -}); - describe('PermissionModeSchema', () => { it('should accept valid permission modes', () => { expect(PermissionModeSchema.parse('default')).toBe('default'); @@ -82,8 +65,8 @@ describe('WorkflowConfigRawSchema', () => { agent: 'coder', allowed_tools: ['Read', 'Grep'], instruction: '{task}', - transitions: [ - { condition: 'done', next_step: 'COMPLETE' }, + rules: [ + { condition: 'Task completed', next: 'COMPLETE' }, ], }, ], @@ -106,8 +89,8 @@ describe('WorkflowConfigRawSchema', () => { allowed_tools: ['Read', 'Edit', 'Write', 'Bash'], permission_mode: 'acceptEdits', instruction: '{task}', - transitions: [ - { condition: 'done', next_step: 'COMPLETE' }, + rules: [ + { condition: 'Done', next: 'COMPLETE' }, ], }, ], @@ -125,7 +108,6 @@ describe('WorkflowConfigRawSchema', () => { name: 'plan', agent: 'planner', instruction: '{task}', - transitions: [], }, ], }; @@ -143,7 +125,6 @@ describe('WorkflowConfigRawSchema', () => { agent: 'coder', permission_mode: 'superAdmin', instruction: '{task}', - transitions: [], }, ], }; @@ -235,31 +216,3 @@ describe('GlobalConfigSchema', () => { expect(result.log_level).toBe('debug'); }); }); - -describe('GENERIC_STATUS_PATTERNS', () => { - it('should have all standard status patterns', () => { - expect(GENERIC_STATUS_PATTERNS.approved).toBeDefined(); - expect(GENERIC_STATUS_PATTERNS.rejected).toBeDefined(); - expect(GENERIC_STATUS_PATTERNS.done).toBeDefined(); - expect(GENERIC_STATUS_PATTERNS.blocked).toBeDefined(); - expect(GENERIC_STATUS_PATTERNS.improve).toBeDefined(); - expect(GENERIC_STATUS_PATTERNS.answer).toBeDefined(); - }); - - it('should have valid regex patterns', () => { - for (const pattern of Object.values(GENERIC_STATUS_PATTERNS)) { - expect(() => new RegExp(pattern)).not.toThrow(); - } - }); - - it('should match any [ROLE:COMMAND] format', () => { - // Generic patterns match any role - expect(new RegExp(GENERIC_STATUS_PATTERNS.approved).test('[CODER:APPROVE]')).toBe(true); - expect(new RegExp(GENERIC_STATUS_PATTERNS.approved).test('[MY_AGENT:APPROVE]')).toBe(true); - expect(new RegExp(GENERIC_STATUS_PATTERNS.done).test('[CUSTOM:DONE]')).toBe(true); - expect(new RegExp(GENERIC_STATUS_PATTERNS.done).test('[CODER:FIXED]')).toBe(true); - expect(new RegExp(GENERIC_STATUS_PATTERNS.improve).test('[MAGI:IMPROVE]')).toBe(true); - expect(new RegExp(GENERIC_STATUS_PATTERNS.answer).test('[PLANNER:ANSWER]')).toBe(true); - expect(new RegExp(GENERIC_STATUS_PATTERNS.answer).test('[MY_AGENT:ANSWER]')).toBe(true); - }); -}); diff --git a/src/__tests__/transitions.test.ts b/src/__tests__/transitions.test.ts new file mode 100644 index 0000000..fcc8890 --- /dev/null +++ b/src/__tests__/transitions.test.ts @@ -0,0 +1,63 @@ +/** + * Tests for workflow transitions module + */ + +import { describe, it, expect } from 'vitest'; +import { determineNextStepByRules } from '../workflow/transitions.js'; +import type { WorkflowStep } from '../models/types.js'; + +function createStepWithRules(rules: { condition: string; next: string }[]): WorkflowStep { + return { + name: 'test-step', + agent: 'test-agent', + agentDisplayName: 'Test Agent', + instructionTemplate: '{task}', + passPreviousResponse: false, + rules: rules.map((r) => ({ + condition: r.condition, + next: r.next, + })), + }; +} + +describe('determineNextStepByRules', () => { + it('should return next step for valid rule index', () => { + const step = createStepWithRules([ + { condition: 'Clear', next: 'implement' }, + { condition: 'Blocked', next: 'ABORT' }, + ]); + + expect(determineNextStepByRules(step, 0)).toBe('implement'); + expect(determineNextStepByRules(step, 1)).toBe('ABORT'); + }); + + it('should return null for out-of-bounds index', () => { + const step = createStepWithRules([ + { condition: 'Clear', next: 'implement' }, + ]); + + expect(determineNextStepByRules(step, 1)).toBeNull(); + expect(determineNextStepByRules(step, -1)).toBeNull(); + expect(determineNextStepByRules(step, 100)).toBeNull(); + }); + + it('should return null when step has no rules', () => { + const step: WorkflowStep = { + name: 'test-step', + agent: 'test-agent', + agentDisplayName: 'Test Agent', + instructionTemplate: '{task}', + passPreviousResponse: false, + }; + + expect(determineNextStepByRules(step, 0)).toBeNull(); + }); + + it('should handle COMPLETE as next step', () => { + const step = createStepWithRules([ + { condition: 'All passed', next: 'COMPLETE' }, + ]); + + expect(determineNextStepByRules(step, 0)).toBe('COMPLETE'); + }); +}); diff --git a/src/agents/runner.ts b/src/agents/runner.ts index c2bcd34..62c4c40 100644 --- a/src/agents/runner.ts +++ b/src/agents/runner.ts @@ -2,7 +2,6 @@ * Agent execution runners */ -import { execSync } from 'node:child_process'; import { existsSync, readFileSync } from 'node:fs'; import { basename, dirname } from 'node:path'; import { @@ -32,8 +31,6 @@ export interface RunAgentOptions { agentPath?: string; /** Allowed tools for this agent run */ allowedTools?: string[]; - /** Status output rules to inject into system prompt */ - statusRulesPrompt?: string; /** Permission mode for tool execution (from workflow step) */ permissionMode?: PermissionMode; onStream?: StreamCallback; @@ -70,28 +67,6 @@ function resolveModel(cwd: string, options?: RunAgentOptions, agentConfig?: Cust return undefined; } -/** Get git diff for review context */ -export function getGitDiff(cwd: string): string { - try { - // First check if HEAD exists (new repos may not have any commits) - try { - execSync('git rev-parse HEAD', { cwd, encoding: 'utf-8', stdio: 'pipe' }); - } catch { - // No commits yet, return empty diff - return ''; - } - - const diff = execSync('git diff HEAD', { - cwd, - encoding: 'utf-8', - maxBuffer: 1024 * 1024 * 10, // 10MB - stdio: 'pipe', - }); - return diff.trim(); - } catch { - return ''; - } -} /** Run a custom agent */ export async function runCustomAgent( @@ -134,12 +109,7 @@ export async function runCustomAgent( } // Custom agent with prompt - let systemPrompt = loadAgentPrompt(agentConfig); - - // Inject status rules if provided - if (options.statusRulesPrompt) { - systemPrompt = `${systemPrompt}\n\n${options.statusRulesPrompt}`; - } + const systemPrompt = loadAgentPrompt(agentConfig); const providerType = resolveProvider(options.cwd, options, agentConfig); const provider = getProvider(providerType); @@ -149,7 +119,6 @@ export async function runCustomAgent( sessionId: options.sessionId, allowedTools, model: resolveModel(options.cwd, options, agentConfig), - statusPatterns: agentConfig.statusPatterns, permissionMode: options.permissionMode, onStream: options.onStream, onPermissionRequest: options.onPermissionRequest, @@ -217,12 +186,7 @@ export async function runAgent( if (!existsSync(options.agentPath)) { throw new Error(`Agent file not found: ${options.agentPath}`); } - let systemPrompt = loadAgentPromptFromPath(options.agentPath); - - // Inject status rules if provided - if (options.statusRulesPrompt) { - systemPrompt = `${systemPrompt}\n\n${options.statusRulesPrompt}`; - } + const systemPrompt = loadAgentPromptFromPath(options.agentPath); const providerType = resolveProvider(options.cwd, options); const provider = getProvider(providerType); diff --git a/src/claude/client.ts b/src/claude/client.ts index c6f9de7..75849f3 100644 --- a/src/claude/client.ts +++ b/src/claude/client.ts @@ -7,7 +7,6 @@ import { executeClaudeCli, type ClaudeSpawnOptions, type StreamCallback, type PermissionHandler, type AskUserQuestionHandler } from './process.js'; import type { AgentDefinition } from '@anthropic-ai/claude-agent-sdk'; import type { AgentResponse, Status, PermissionMode } from '../models/types.js'; -import { GENERIC_STATUS_PATTERNS } from '../models/schemas.js'; import { createLogger } from '../utils/debug.js'; const log = createLogger('client'); @@ -20,7 +19,6 @@ export interface ClaudeCallOptions { model?: string; maxTurns?: number; systemPrompt?: string; - statusPatterns?: Record; /** SDK agents to register for sub-agent execution */ agents?: Record; /** Permission mode for tool execution (from workflow step) */ @@ -35,22 +33,21 @@ export interface ClaudeCallOptions { bypassPermissions?: boolean; } -/** Detect status from agent output content */ -export function detectStatus( - content: string, - patterns: Record -): Status { - for (const [status, pattern] of Object.entries(patterns)) { - try { - const regex = new RegExp(pattern, 'i'); - if (regex.test(content)) { - return status as Status; - } - } catch { - // Invalid regex, skip - } +/** + * Detect rule index from numbered tag pattern [STEP_NAME:N]. + * Returns 0-based rule index, or -1 if no match. + * + * Example: detectRuleIndex("... [PLAN:2] ...", "plan") → 1 + */ +export function detectRuleIndex(content: string, stepName: string): number { + const tag = stepName.toUpperCase(); + const regex = new RegExp(`\\[${tag}:(\\d+)\\]`, 'i'); + const match = content.match(regex); + if (match?.[1]) { + const index = Number.parseInt(match[1], 10) - 1; + return index >= 0 ? index : -1; } - return 'in_progress'; + return -1; } /** Validate regex pattern for ReDoS safety */ @@ -79,28 +76,17 @@ export function isRegexSafe(pattern: string): boolean { return true; } -/** Get status patterns for a built-in agent type */ -export function getBuiltinStatusPatterns(_agentType: string): Record { - // Uses generic patterns that work for any agent - return GENERIC_STATUS_PATTERNS; -} - /** Determine status from result */ function determineStatus( result: { success: boolean; interrupted?: boolean; content: string; fullContent?: string }, - patterns: Record ): Status { if (!result.success) { - // Check if it was an interrupt using the flag (not magic string) if (result.interrupted) { return 'interrupted'; } return 'blocked'; } - // Use fullContent for status detection (contains all assistant messages) - // Fall back to content if fullContent is not available - const textForStatusDetection = result.fullContent || result.content; - return detectStatus(textForStatusDetection, patterns); + return 'done'; } /** Call Claude with an agent prompt */ @@ -125,8 +111,7 @@ export async function callClaude( }; const result = await executeClaudeCli(prompt, spawnOptions); - const patterns = options.statusPatterns || getBuiltinStatusPatterns(agentType); - const status = determineStatus(result, patterns); + const status = determineStatus(result); if (!result.success && result.error) { log.error('Agent query failed', { agent: agentType, error: result.error }); @@ -164,9 +149,7 @@ export async function callClaudeCustom( }; const result = await executeClaudeCli(prompt, spawnOptions); - // Use provided patterns, or fall back to built-in patterns for known agents - const patterns = options.statusPatterns || getBuiltinStatusPatterns(agentName); - const status = determineStatus(result, patterns); + const status = determineStatus(result); if (!result.success && result.error) { log.error('Agent query failed', { agent: agentName, error: result.error }); diff --git a/src/claude/index.ts b/src/claude/index.ts index 83be70e..bfafe55 100644 --- a/src/claude/index.ts +++ b/src/claude/index.ts @@ -57,8 +57,7 @@ export { callClaudeCustom, callClaudeAgent, callClaudeSkill, - detectStatus, + detectRuleIndex, isRegexSafe, - getBuiltinStatusPatterns, type ClaudeCallOptions, } from './client.js'; diff --git a/src/codex/client.ts b/src/codex/client.ts index 8412904..9ae70aa 100644 --- a/src/codex/client.ts +++ b/src/codex/client.ts @@ -6,8 +6,6 @@ import { Codex } from '@openai/codex-sdk'; import type { AgentResponse, Status } from '../models/types.js'; -import { GENERIC_STATUS_PATTERNS } from '../models/schemas.js'; -import { detectStatus } from '../claude/client.js'; import type { StreamCallback } from '../claude/process.js'; import { createLogger } from '../utils/debug.js'; @@ -19,7 +17,6 @@ export interface CodexCallOptions { sessionId?: string; model?: string; systemPrompt?: string; - statusPatterns?: Record; /** Enable streaming mode with callback (best-effort) */ onStream?: StreamCallback; } @@ -101,8 +98,8 @@ function emitResult( }); } -function determineStatus(content: string, patterns: Record): Status { - return detectStatus(content, patterns); +function determineStatus(success: boolean): Status { + return success ? 'done' : 'blocked'; } type CodexEvent = { @@ -476,8 +473,7 @@ export async function callCodex( const trimmed = content.trim(); emitResult(options.onStream, true, trimmed, threadId); - const patterns = options.statusPatterns || GENERIC_STATUS_PATTERNS; - const status = determineStatus(trimmed, patterns); + const status = determineStatus(true); return { agent: agentType, diff --git a/src/commands/workflowExecution.ts b/src/commands/workflowExecution.ts index b5707e7..0c7be9f 100644 --- a/src/commands/workflowExecution.ts +++ b/src/commands/workflowExecution.ts @@ -2,6 +2,7 @@ * Workflow execution logic */ +import { readFileSync } from 'node:fs'; import { WorkflowEngine } from '../workflow/engine.js'; import type { WorkflowConfig, Language } from '../models/types.js'; import type { IterationLimitRequest } from '../workflow/types.js'; @@ -184,10 +185,11 @@ export async function executeWorkflow( displayRef.current = new StreamDisplay(step.agentDisplayName); }); - engine.on('step:complete', (step, response) => { + engine.on('step:complete', (step, response, instruction) => { log.debug('Step completed', { step: step.name, status: response.status, + matchedRuleIndex: response.matchedRuleIndex, contentLength: response.content.length, sessionId: response.sessionId, error: response.error, @@ -197,20 +199,37 @@ export async function executeWorkflow( displayRef.current = null; } console.log(); - status('Status', response.status); + + if (response.matchedRuleIndex != null && step.rules) { + const rule = step.rules[response.matchedRuleIndex]; + if (rule) { + status('Status', rule.condition); + } else { + status('Status', response.status); + } + } else { + status('Status', response.status); + } + if (response.error) { error(`Error: ${response.error}`); } if (response.sessionId) { status('Session', response.sessionId); } - addToSessionLog(sessionLog, step.name, response); + addToSessionLog(sessionLog, step.name, response, instruction); // Incremental save after each step saveSessionLog(sessionLog, workflowSessionId, projectCwd); updateLatestPointer(sessionLog, workflowSessionId, projectCwd); }); + engine.on('step:report', (_step, filePath, fileName) => { + const content = readFileSync(filePath, 'utf-8'); + console.log(`\n📄 Report: ${fileName}\n`); + console.log(content); + }); + engine.on('workflow:complete', (state) => { log.info('Workflow completed successfully', { iterations: state.iteration }); finalizeSessionLog(sessionLog, 'completed'); diff --git a/src/config/workflowLoader.ts b/src/config/workflowLoader.ts index adc6f0d..93c9f9a 100644 --- a/src/config/workflowLoader.ts +++ b/src/config/workflowLoader.ts @@ -8,7 +8,7 @@ import { readFileSync, existsSync, readdirSync, statSync } from 'node:fs'; import { join, dirname, basename } from 'node:path'; import { parse as parseYaml } from 'yaml'; import { WorkflowConfigRawSchema } from '../models/schemas.js'; -import type { WorkflowConfig, WorkflowStep } from '../models/types.js'; +import type { WorkflowConfig, WorkflowStep, WorkflowRule, ReportConfig, ReportObjectConfig } from '../models/types.js'; import { getGlobalWorkflowsDir } from './paths.js'; /** Get builtin workflow by name */ @@ -54,6 +54,72 @@ function extractAgentDisplayName(agentPath: string): string { return filename; } +/** + * Resolve a string value that may be a file path. + * If the value ends with .md and the file exists (resolved relative to workflowDir), + * read and return the file contents. Otherwise return the value as-is. + */ +function resolveContentPath(value: string | undefined, workflowDir: string): string | undefined { + if (value == null) return undefined; + if (value.endsWith('.md')) { + // Resolve path relative to workflow directory + let resolvedPath = value; + if (value.startsWith('./')) { + resolvedPath = join(workflowDir, value.slice(2)); + } else if (value.startsWith('~')) { + const homedir = process.env.HOME || process.env.USERPROFILE || ''; + resolvedPath = join(homedir, value.slice(1)); + } else if (!value.startsWith('/')) { + resolvedPath = join(workflowDir, value); + } + if (existsSync(resolvedPath)) { + return readFileSync(resolvedPath, 'utf-8'); + } + } + return value; +} + +/** + * Check if a raw report value is the object form (has 'name' property). + */ +function isReportObject(raw: unknown): raw is { name: string; order?: string; format?: string } { + return typeof raw === 'object' && raw !== null && !Array.isArray(raw) && 'name' in raw; +} + +/** + * Normalize the raw report field from YAML into internal format. + * + * YAML formats: + * report: "00-plan.md" → string (single file) + * report: → ReportConfig[] (multiple files) + * - Scope: 01-scope.md + * - Decisions: 02-decisions.md + * report: → ReportObjectConfig (object form) + * name: 00-plan.md + * order: ... + * format: ... + * + * Array items are parsed as single-key objects: [{Scope: "01-scope.md"}, ...] + */ +function normalizeReport( + raw: string | Record[] | { name: string; order?: string; format?: string } | undefined, + workflowDir: string, +): string | ReportConfig[] | ReportObjectConfig | undefined { + if (raw == null) return undefined; + if (typeof raw === 'string') return raw; + if (isReportObject(raw)) { + return { + name: raw.name, + order: resolveContentPath(raw.order, workflowDir), + format: resolveContentPath(raw.format, workflowDir), + }; + } + // Convert [{Scope: "01-scope.md"}, ...] to [{label: "Scope", path: "01-scope.md"}, ...] + return (raw as Record[]).flatMap((entry) => + Object.entries(entry).map(([label, path]) => ({ label, path })), + ); +} + /** * Convert raw YAML workflow config to internal format. * Agent paths are resolved relative to the workflow directory. @@ -61,24 +127,29 @@ function extractAgentDisplayName(agentPath: string): string { function normalizeWorkflowConfig(raw: unknown, workflowDir: string): WorkflowConfig { const parsed = WorkflowConfigRawSchema.parse(raw); - const steps: WorkflowStep[] = parsed.steps.map((step) => ({ - name: step.name, - agent: step.agent, - agentDisplayName: step.agent_name || extractAgentDisplayName(step.agent), - agentPath: resolveAgentPathForWorkflow(step.agent, workflowDir), - allowedTools: step.allowed_tools, - provider: step.provider, - model: step.model, - permissionMode: step.permission_mode, - instructionTemplate: step.instruction_template || step.instruction || '{task}', - statusRulesPrompt: step.status_rules_prompt, - transitions: step.transitions.map((t) => ({ - condition: t.condition, - nextStep: t.next_step, - })), - passPreviousResponse: step.pass_previous_response, - onNoStatus: step.on_no_status, - })); + const steps: WorkflowStep[] = parsed.steps.map((step) => { + const rules: WorkflowRule[] | undefined = step.rules?.map((r) => ({ + condition: r.condition, + next: r.next, + appendix: r.appendix, + })); + + return { + name: step.name, + agent: step.agent, + agentDisplayName: step.agent_name || extractAgentDisplayName(step.agent), + agentPath: resolveAgentPathForWorkflow(step.agent, workflowDir), + allowedTools: step.allowed_tools, + provider: step.provider, + model: step.model, + permissionMode: step.permission_mode, + edit: step.edit, + instructionTemplate: resolveContentPath(step.instruction_template, workflowDir) || step.instruction || '{task}', + rules, + report: normalizeReport(step.report, workflowDir), + passPreviousResponse: step.pass_previous_response, + }; + }); return { name: parsed.name, diff --git a/src/models/index.ts b/src/models/index.ts index 430d792..5b3e865 100644 --- a/src/models/index.ts +++ b/src/models/index.ts @@ -2,10 +2,10 @@ export type { AgentType, Status, - TransitionCondition, + ReportConfig, + ReportObjectConfig, AgentResponse, SessionState, - WorkflowTransition, WorkflowStep, WorkflowConfig, WorkflowState, diff --git a/src/models/schemas.ts b/src/models/schemas.ts index e0379f1..e1d7aab 100644 --- a/src/models/schemas.ts +++ b/src/models/schemas.ts @@ -24,41 +24,63 @@ export const StatusSchema = z.enum([ 'answer', ]); -/** - * Transition condition schema - * - * WARNING: Do NOT add new values carelessly. - * Use existing values creatively in workflow design: - * - done: Task completed (minor fixes, successful completion) - * - blocked: Cannot proceed (needs plan rework) - * - approved: Review passed - * - rejected: Review failed, needs major rework - * - improve: Needs improvement (security concerns, quality issues) - * - answer: Question answered (complete workflow as success) - * - always: Unconditional transition - */ -export const TransitionConditionSchema = z.enum([ - 'done', - 'blocked', - 'approved', - 'rejected', - 'improve', - 'answer', - 'always', -]); - -/** On no status behavior schema */ -export const OnNoStatusBehaviorSchema = z.enum(['complete', 'continue', 'stay']); - -/** Workflow transition schema */ -export const WorkflowTransitionSchema = z.object({ - condition: TransitionConditionSchema, - nextStep: z.string().min(1), -}); - /** Permission mode schema for tool execution */ export const PermissionModeSchema = z.enum(['default', 'acceptEdits', 'bypassPermissions']); +/** + * Report object schema (new structured format). + * + * YAML format: + * report: + * name: 00-plan.md + * order: | + * **レポート出力:** {report:00-plan.md} に出力してください。 + * format: | + * **レポートフォーマット:** + * ```markdown + * ... + * ``` + */ +export const ReportObjectSchema = z.object({ + /** Report file name */ + name: z.string().min(1), + /** Instruction prepended before instruction_template (e.g., output destination) */ + order: z.string().optional(), + /** Instruction appended after instruction_template (e.g., output format) */ + format: z.string().optional(), +}); + +/** + * Report field schema. + * + * YAML formats: + * report: 00-plan.md # single file (string) + * report: # multiple files (label: path map entries) + * - Scope: 01-scope.md + * - Decisions: 02-decisions.md + * report: # object form (name + order + format) + * name: 00-plan.md + * order: ... + * format: ... + * + * Array items are parsed as single-key objects: [{Scope: "01-scope.md"}, ...] + */ +export const ReportFieldSchema = z.union([ + z.string().min(1), + z.array(z.record(z.string(), z.string())).min(1), + ReportObjectSchema, +]); + +/** Rule-based transition schema (new unified format) */ +export const WorkflowRuleSchema = z.object({ + /** Human-readable condition text */ + condition: z.string().min(1), + /** Next step name (e.g., implement, COMPLETE, ABORT) */ + next: z.string().min(1), + /** Template for additional AI output */ + appendix: z.string().optional(), +}); + /** Workflow step schema - raw YAML format */ export const WorkflowStepRawSchema = z.object({ name: z.string().min(1), @@ -70,17 +92,15 @@ export const WorkflowStepRawSchema = z.object({ model: z.string().optional(), /** Permission mode for tool execution in this step */ permission_mode: PermissionModeSchema.optional(), + /** Whether this step is allowed to edit project files */ + edit: z.boolean().optional(), instruction: z.string().optional(), instruction_template: z.string().optional(), - status_rules_prompt: z.string().optional(), + /** Rules for step routing */ + rules: z.array(WorkflowRuleSchema).optional(), + /** Report file(s) for this step */ + report: ReportFieldSchema.optional(), pass_previous_response: z.boolean().optional().default(true), - on_no_status: OnNoStatusBehaviorSchema.optional(), - transitions: z.array( - z.object({ - condition: TransitionConditionSchema, - next_step: z.string().min(1), - }) - ).optional().default([]), }); /** Workflow configuration schema - raw YAML format */ @@ -99,7 +119,6 @@ export const CustomAgentConfigSchema = z.object({ prompt_file: z.string().optional(), prompt: z.string().optional(), allowed_tools: z.array(z.string()).optional(), - status_patterns: z.record(z.string(), z.string()).optional(), claude_agent: z.string().optional(), claude_skill: z.string().optional(), provider: z.enum(['claude', 'codex', 'mock']).optional(), @@ -138,18 +157,3 @@ export const ProjectConfigSchema = z.object({ provider: z.enum(['claude', 'codex', 'mock']).optional(), }); -/** - * Generic status patterns that match any role name - * Format: [ROLE:COMMAND] where ROLE is any word characters - * - * This allows new agents to be added without modifying this file. - * Custom agents can override these patterns in their configuration. - */ -export const GENERIC_STATUS_PATTERNS: Record = { - approved: '\\[[\\w-]+:APPROVE\\]', - rejected: '\\[[\\w-]+:REJECT\\]', - improve: '\\[[\\w-]+:IMPROVE\\]', - done: '\\[[\\w-]+:(DONE|FIXED)\\]', - blocked: '\\[[\\w-]+:BLOCKED\\]', - answer: '\\[[\\w-]+:ANSWER\\]', -}; diff --git a/src/models/types.ts b/src/models/types.ts index 53ffe30..8bce74c 100644 --- a/src/models/types.ts +++ b/src/models/types.ts @@ -18,16 +18,6 @@ export type Status = | 'interrupted' | 'answer'; -/** Condition types for workflow transitions */ -export type TransitionCondition = - | 'done' - | 'blocked' - | 'approved' - | 'rejected' - | 'improve' - | 'answer' - | 'always'; - /** Response from an agent execution */ export interface AgentResponse { agent: string; @@ -37,6 +27,8 @@ export interface AgentResponse { sessionId?: string; /** Error message when the query failed (e.g., API error, rate limit) */ error?: string; + /** Matched rule index (0-based) when rules-based detection was used */ + matchedRuleIndex?: number; } /** Session state for workflow execution */ @@ -48,14 +40,33 @@ export interface SessionState { context: Record; } -/** Workflow step transition configuration */ -export interface WorkflowTransition { - condition: TransitionCondition; - nextStep: string; +/** Rule-based transition configuration (new unified format) */ +export interface WorkflowRule { + /** Human-readable condition text */ + condition: string; + /** Next step name (e.g., implement, COMPLETE, ABORT) */ + next: string; + /** Template for additional AI output */ + appendix?: string; } -/** Behavior when no status marker is found in agent output */ -export type OnNoStatusBehavior = 'complete' | 'continue' | 'stay'; +/** Report file configuration for a workflow step (label: path pair) */ +export interface ReportConfig { + /** Display label (e.g., "Scope", "Decisions") */ + label: string; + /** File path relative to report directory (e.g., "01-coder-scope.md") */ + path: string; +} + +/** Report object configuration with order/format instructions */ +export interface ReportObjectConfig { + /** Report file name (e.g., "00-plan.md") */ + name: string; + /** Instruction prepended before instruction_template (e.g., output destination) */ + order?: string; + /** Instruction appended after instruction_template (e.g., output format) */ + format?: string; +} /** Permission mode for tool execution */ export type PermissionMode = 'default' | 'acceptEdits' | 'bypassPermissions'; @@ -77,18 +88,14 @@ export interface WorkflowStep { model?: string; /** Permission mode for tool execution in this step */ permissionMode?: PermissionMode; + /** Whether this step is allowed to edit project files (true=allowed, false=prohibited, undefined=no prompt) */ + edit?: boolean; instructionTemplate: string; - /** Status output rules to be injected into system prompt */ - statusRulesPrompt?: string; - transitions: WorkflowTransition[]; + /** Rules for step routing */ + rules?: WorkflowRule[]; + /** Report file configuration. Single string, array of label:path, or object with order/format. */ + report?: string | ReportConfig[] | ReportObjectConfig; passPreviousResponse: boolean; - /** - * Behavior when agent doesn't output a status marker (in_progress). - * - 'complete': Treat as done, follow done/always transition or complete workflow (default) - * - 'continue': Treat as done, follow done/always transition or move to next step - * - 'stay': Stay on current step (may cause loops, use with caution) - */ - onNoStatus?: OnNoStatusBehavior; } /** Loop detection configuration */ @@ -135,7 +142,6 @@ export interface CustomAgentConfig { promptFile?: string; prompt?: string; allowedTools?: string[]; - statusPatterns?: Record; claudeAgent?: string; claudeSkill?: string; provider?: 'claude' | 'codex' | 'mock'; diff --git a/src/providers/claude.ts b/src/providers/claude.ts index e5fdc4d..e60ef74 100644 --- a/src/providers/claude.ts +++ b/src/providers/claude.ts @@ -15,7 +15,6 @@ export class ClaudeProvider implements Provider { allowedTools: options.allowedTools, model: options.model, systemPrompt: options.systemPrompt, - statusPatterns: options.statusPatterns, permissionMode: options.permissionMode, onStream: options.onStream, onPermissionRequest: options.onPermissionRequest, @@ -32,7 +31,6 @@ export class ClaudeProvider implements Provider { sessionId: options.sessionId, allowedTools: options.allowedTools, model: options.model, - statusPatterns: options.statusPatterns, permissionMode: options.permissionMode, onStream: options.onStream, onPermissionRequest: options.onPermissionRequest, diff --git a/src/providers/codex.ts b/src/providers/codex.ts index 408956f..119fb99 100644 --- a/src/providers/codex.ts +++ b/src/providers/codex.ts @@ -14,7 +14,6 @@ export class CodexProvider implements Provider { sessionId: options.sessionId, model: options.model, systemPrompt: options.systemPrompt, - statusPatterns: options.statusPatterns, onStream: options.onStream, }; @@ -26,7 +25,6 @@ export class CodexProvider implements Provider { cwd: options.cwd, sessionId: options.sessionId, model: options.model, - statusPatterns: options.statusPatterns, onStream: options.onStream, }; diff --git a/src/providers/index.ts b/src/providers/index.ts index 56c31ec..c8d9c7f 100644 --- a/src/providers/index.ts +++ b/src/providers/index.ts @@ -18,7 +18,6 @@ export interface ProviderCallOptions { model?: string; systemPrompt?: string; allowedTools?: string[]; - statusPatterns?: Record; /** Permission mode for tool execution (from workflow step) */ permissionMode?: PermissionMode; onStream?: StreamCallback; diff --git a/src/utils/session.ts b/src/utils/session.ts index 27d5b6b..fa88f27 100644 --- a/src/utils/session.ts +++ b/src/utils/session.ts @@ -19,6 +19,7 @@ export interface SessionLog { history: Array<{ step: string; agent: string; + instruction: string; status: string; timestamp: string; content: string; @@ -76,11 +77,13 @@ export function createSessionLog( export function addToSessionLog( log: SessionLog, stepName: string, - response: AgentResponse + response: AgentResponse, + instruction: string ): void { log.history.push({ step: stepName, agent: response.agent, + instruction, status: response.status, timestamp: response.timestamp.toISOString(), content: response.content, @@ -214,6 +217,7 @@ export function workflowStateToSessionLog( log.history.push({ step: stepName, agent: response.agent, + instruction: '', status: response.status, timestamp: response.timestamp.toISOString(), content: response.content, diff --git a/src/workflow/engine.ts b/src/workflow/engine.ts index ae61595..6e311e8 100644 --- a/src/workflow/engine.ts +++ b/src/workflow/engine.ts @@ -14,8 +14,9 @@ import type { import { runAgent, type RunAgentOptions } from '../agents/runner.js'; import { COMPLETE_STEP, ABORT_STEP, ERROR_MESSAGES } from './constants.js'; import type { WorkflowEngineOptions } from './types.js'; -import { determineNextStep } from './transitions.js'; -import { buildInstruction as buildInstructionFromTemplate } from './instruction-builder.js'; +import { determineNextStepByRules } from './transitions.js'; +import { detectRuleIndex } from '../claude/client.js'; +import { buildInstruction as buildInstructionFromTemplate, isReportObjectConfig } from './instruction-builder.js'; import { LoopDetector } from './loop-detector.js'; import { handleBlocked } from './blocked-handler.js'; import { @@ -105,11 +106,13 @@ export class WorkflowEngine extends EventEmitter { stepNames.add(ABORT_STEP); for (const step of this.config.steps) { - for (const transition of step.transitions) { - if (!stepNames.has(transition.nextStep)) { - throw new Error( - `Invalid transition in step "${step.name}": target step "${transition.nextStep}" does not exist` - ); + if (step.rules) { + for (const rule of step.rules) { + if (!stepNames.has(rule.next)) { + throw new Error( + `Invalid rule in step "${step.name}": target step "${rule.next}" does not exist` + ); + } } } } @@ -165,9 +168,36 @@ export class WorkflowEngine extends EventEmitter { return step; } + /** + * Emit step:report events for each report file that exists after step completion. + * The UI layer (workflowExecution.ts) listens and displays the content. + */ + private emitStepReports(step: WorkflowStep): void { + if (!step.report || !this.reportDir) return; + const baseDir = join(this.projectCwd, '.takt', 'reports', this.reportDir); + + if (typeof step.report === 'string') { + this.emitIfReportExists(step, baseDir, step.report); + } else if (isReportObjectConfig(step.report)) { + this.emitIfReportExists(step, baseDir, step.report.name); + } else { + // ReportConfig[] (array) + for (const rc of step.report) { + this.emitIfReportExists(step, baseDir, rc.path); + } + } + } + + /** Emit step:report if the report file exists */ + private emitIfReportExists(step: WorkflowStep, baseDir: string, fileName: string): void { + const filePath = join(baseDir, fileName); + if (existsSync(filePath)) { + this.emit('step:report', step, filePath, fileName); + } + } + /** Run a single step */ - private async runStep(step: WorkflowStep): Promise { - // Increment step iteration counter before building instruction + private async runStep(step: WorkflowStep): Promise<{ response: AgentResponse; instruction: string }> { const stepIteration = incrementStepIteration(this.state, step.name); const instruction = this.buildInstruction(step, stepIteration); const sessionId = this.state.agentSessions.get(step.agent); @@ -184,7 +214,6 @@ export class WorkflowEngine extends EventEmitter { sessionId, agentPath: step.agentPath, allowedTools: step.allowedTools, - statusRulesPrompt: step.statusRulesPrompt, provider: step.provider, model: step.model, permissionMode: step.permissionMode, @@ -194,7 +223,7 @@ export class WorkflowEngine extends EventEmitter { bypassPermissions: this.options.bypassPermissions, }; - const response = await runAgent(step.agent, instruction, agentOptions); + let response = await runAgent(step.agent, instruction, agentOptions); if (response.sessionId) { const previousSessionId = this.state.agentSessions.get(step.agent); @@ -205,8 +234,30 @@ export class WorkflowEngine extends EventEmitter { } } + if (step.rules && step.rules.length > 0) { + const ruleIndex = detectRuleIndex(response.content, step.name); + if (ruleIndex >= 0 && ruleIndex < step.rules.length) { + response = { ...response, matchedRuleIndex: ruleIndex }; + } + } + this.state.stepOutputs.set(step.name, response); - return response; + this.emitStepReports(step); + return { response, instruction }; + } + + /** + * Determine next step for a completed step using rules-based routing. + */ + private resolveNextStep(step: WorkflowStep, response: AgentResponse): string { + if (response.matchedRuleIndex != null && step.rules) { + const nextByRules = determineNextStepByRules(step, response.matchedRuleIndex); + if (nextByRules) { + return nextByRules; + } + } + + throw new Error(`No matching rule found for step "${step.name}" (status: ${response.status})`); } /** Run the workflow to completion */ @@ -253,8 +304,8 @@ export class WorkflowEngine extends EventEmitter { this.emit('step:start', step, this.state.iteration); try { - const response = await this.runStep(step); - this.emit('step:complete', step, response); + const { response, instruction } = await this.runStep(step); + this.emit('step:complete', step, response, instruction); if (response.status === 'blocked') { this.emit('step:blocked', step, response); @@ -271,10 +322,11 @@ export class WorkflowEngine extends EventEmitter { break; } - const nextStep = determineNextStep(step, response.status, this.config); + const nextStep = this.resolveNextStep(step, response); log.debug('Step transition', { from: step.name, status: response.status, + matchedRuleIndex: response.matchedRuleIndex, nextStep, }); @@ -328,8 +380,8 @@ export class WorkflowEngine extends EventEmitter { } this.state.iteration++; - const response = await this.runStep(step); - const nextStep = determineNextStep(step, response.status, this.config); + const { response } = await this.runStep(step); + const nextStep = this.resolveNextStep(step, response); const isComplete = nextStep === COMPLETE_STEP || nextStep === ABORT_STEP; if (!isComplete) { diff --git a/src/workflow/index.ts b/src/workflow/index.ts index 92fc6ab..b66cfb8 100644 --- a/src/workflow/index.ts +++ b/src/workflow/index.ts @@ -23,7 +23,7 @@ export type { } from './types.js'; // Transitions -export { determineNextStep, matchesCondition, extractBlockedPrompt } from './transitions.js'; +export { determineNextStepByRules, extractBlockedPrompt } from './transitions.js'; // Loop detection export { LoopDetector } from './loop-detector.js'; diff --git a/src/workflow/instruction-builder.ts b/src/workflow/instruction-builder.ts index 72bee38..cb952c5 100644 --- a/src/workflow/instruction-builder.ts +++ b/src/workflow/instruction-builder.ts @@ -1,12 +1,15 @@ /** * Instruction template builder for workflow steps * - * Builds the instruction string for agent execution by replacing - * template placeholders with actual values. + * Builds the instruction string for agent execution by: + * 1. Auto-injecting standard sections (Execution Context, Workflow Context, + * User Request, Previous Response, Additional User Inputs, Instructions header) + * 2. Replacing template placeholders with actual values + * 3. Appending auto-generated status rules from workflow rules */ -import type { WorkflowStep, AgentResponse, Language } from '../models/types.js'; -import { getGitDiff } from '../agents/runner.js'; +import type { WorkflowStep, WorkflowRule, AgentResponse, Language, ReportConfig, ReportObjectConfig } from '../models/types.js'; + /** * Context for building instruction from template. @@ -40,17 +43,20 @@ export interface ExecutionMetadata { readonly workingDirectory: string; /** Language for metadata rendering */ readonly language: Language; + /** Whether file editing is allowed for this step (undefined = no prompt) */ + readonly edit?: boolean; } /** - * Build execution metadata from instruction context. + * Build execution metadata from instruction context and step config. * - * Pure function: InstructionContext → ExecutionMetadata. + * Pure function: (InstructionContext, edit?) → ExecutionMetadata. */ -export function buildExecutionMetadata(context: InstructionContext): ExecutionMetadata { +export function buildExecutionMetadata(context: InstructionContext, edit?: boolean): ExecutionMetadata { return { workingDirectory: context.cwd, language: context.language ?? 'en', + edit, }; } @@ -70,13 +76,97 @@ const STATUS_RULES_HEADER_STRINGS = { /** * Render status rules header. - * Prepended to status_rules_prompt when it exists. + * Prepended to auto-generated status rules from workflow rules. */ export function renderStatusRulesHeader(language: Language): string { const strings = STATUS_RULES_HEADER_STRINGS[language]; return [strings.heading, '', strings.warning, strings.instruction, ''].join('\n'); } +/** Localized strings for rules-based status prompt */ +const RULES_PROMPT_STRINGS = { + en: { + criteriaHeading: '## Decision Criteria', + headerNum: '#', + headerCondition: 'Condition', + headerTag: 'Tag', + outputHeading: '## Output Format', + outputInstruction: 'Output the tag corresponding to your decision:', + appendixHeading: '### Appendix Template', + appendixInstruction: 'When outputting `[{tag}]`, append the following:', + }, + ja: { + criteriaHeading: '## 判定基準', + headerNum: '#', + headerCondition: '状況', + headerTag: 'タグ', + outputHeading: '## 出力フォーマット', + outputInstruction: '判定に対応するタグを出力してください:', + appendixHeading: '### 追加出力テンプレート', + appendixInstruction: '`[{tag}]` を出力する場合、以下を追記してください:', + }, +} as const; + +/** + * Generate status rules prompt from rules configuration. + * Creates a structured prompt that tells the agent which numbered tags to output. + * + * Example output for step "plan" with 3 rules: + * ## 判定基準 + * | # | 状況 | タグ | + * |---|------|------| + * | 1 | 要件が明確で実装可能 | `[PLAN:1]` | + * | 2 | ユーザーが質問をしている | `[PLAN:2]` | + * | 3 | 要件が不明確、情報不足 | `[PLAN:3]` | + */ +export function generateStatusRulesFromRules( + stepName: string, + rules: WorkflowRule[], + language: Language, +): string { + const tag = stepName.toUpperCase(); + const strings = RULES_PROMPT_STRINGS[language]; + + const lines: string[] = []; + + // Criteria table + lines.push(strings.criteriaHeading); + lines.push(''); + lines.push(`| ${strings.headerNum} | ${strings.headerCondition} | ${strings.headerTag} |`); + lines.push('|---|------|------|'); + for (const [i, rule] of rules.entries()) { + lines.push(`| ${i + 1} | ${rule.condition} | \`[${tag}:${i + 1}]\` |`); + } + lines.push(''); + + // Output format + lines.push(strings.outputHeading); + lines.push(''); + lines.push(strings.outputInstruction); + lines.push(''); + for (const [i, rule] of rules.entries()) { + lines.push(`- \`[${tag}:${i + 1}]\` — ${rule.condition}`); + } + + // Appendix templates (if any rules have appendix) + const rulesWithAppendix = rules.filter((r) => r.appendix); + if (rulesWithAppendix.length > 0) { + lines.push(''); + lines.push(strings.appendixHeading); + for (const [i, rule] of rules.entries()) { + if (!rule.appendix) continue; + const tagStr = `[${tag}:${i + 1}]`; + lines.push(''); + lines.push(strings.appendixInstruction.replace('{tag}', tagStr)); + lines.push('```'); + lines.push(rule.appendix.trimEnd()); + lines.push('```'); + } + } + + return lines.join('\n'); +} + /** Localized strings for execution metadata rendering */ const METADATA_STRINGS = { en: { @@ -85,6 +175,8 @@ const METADATA_STRINGS = { rulesHeading: '## Execution Rules', noCommit: '**Do NOT run git commit.** Commits are handled automatically by the system after workflow completion.', noCd: '**Do NOT use `cd` in Bash commands.** Your working directory is already set correctly. Run commands directly without changing directories.', + editEnabled: '**Editing is ENABLED for this step.** You may create, modify, and delete files as needed to fulfill the user\'s request.', + editDisabled: '**Editing is DISABLED for this step.** Do NOT create, modify, or delete any project source files. You may only read/search code and write to report files in the Report Directory.', note: 'Note: This section is metadata. Follow the language used in the rest of the prompt.', }, ja: { @@ -93,6 +185,8 @@ const METADATA_STRINGS = { rulesHeading: '## 実行ルール', noCommit: '**git commit を実行しないでください。** コミットはワークフロー完了後にシステムが自動で行います。', noCd: '**Bashコマンドで `cd` を使用しないでください。** 作業ディレクトリは既に正しく設定されています。ディレクトリを変更せずにコマンドを実行してください。', + editEnabled: '**このステップでは編集が許可されています。** ユーザーの要求に応じて、ファイルの作成・変更・削除を行ってください。', + editDisabled: '**このステップでは編集が禁止されています。** プロジェクトのソースファイルを作成・変更・削除しないでください。コードの読み取り・検索と、Report Directoryへのレポート出力のみ行えます。', note: '', }, } as const; @@ -114,6 +208,11 @@ export function renderExecutionMetadata(metadata: ExecutionMetadata): string { `- ${strings.noCommit}`, `- ${strings.noCd}`, ]; + if (metadata.edit === true) { + lines.push(`- ${strings.editEnabled}`); + } else if (metadata.edit === false) { + lines.push(`- ${strings.editDisabled}`); + } if (strings.note) { lines.push(''); lines.push(strings.note); @@ -130,73 +229,276 @@ function escapeTemplateChars(str: string): string { } /** - * Build instruction from template with context values. - * - * Supported placeholders: - * - {task} - The main task/prompt - * - {iteration} - Current iteration number (workflow-wide turn count) - * - {max_iterations} - Maximum iterations allowed - * - {step_iteration} - Current step's iteration number (how many times this step has been executed) - * - {previous_response} - Output from previous step (if passPreviousResponse is true) - * - {git_diff} - Current git diff output - * - {user_inputs} - Accumulated user inputs - * - {report_dir} - Report directory name (e.g., "20250126-143052-task-summary") + * Check if a report config is the object form (ReportObjectConfig). */ -export function buildInstruction( +export function isReportObjectConfig(report: string | ReportConfig[] | ReportObjectConfig): report is ReportObjectConfig { + return typeof report === 'object' && !Array.isArray(report) && 'name' in report; +} + +/** Localized strings for auto-injected sections */ +const SECTION_STRINGS = { + en: { + workflowContext: '## Workflow Context', + iteration: 'Iteration', + iterationWorkflowWide: '(workflow-wide)', + stepIteration: 'Step Iteration', + stepIterationTimes: '(times this step has run)', + step: 'Step', + reportDirectory: 'Report Directory', + reportFile: 'Report File', + reportFiles: 'Report Files', + userRequest: '## User Request', + previousResponse: '## Previous Response', + additionalUserInputs: '## Additional User Inputs', + instructions: '## Instructions', + }, + ja: { + workflowContext: '## Workflow Context', + iteration: 'Iteration', + iterationWorkflowWide: '（ワークフロー全体）', + stepIteration: 'Step Iteration', + stepIterationTimes: '（このステップの実行回数）', + step: 'Step', + reportDirectory: 'Report Directory', + reportFile: 'Report File', + reportFiles: 'Report Files', + userRequest: '## User Request', + previousResponse: '## Previous Response', + additionalUserInputs: '## Additional User Inputs', + instructions: '## Instructions', + }, +} as const; + +/** Localized strings for auto-generated report output instructions */ +const REPORT_OUTPUT_STRINGS = { + en: { + singleHeading: '**Report output:** Output to the `Report File` specified above.', + multiHeading: '**Report output:** Output to the `Report Files` specified above.', + createRule: '- If file does not exist: Create new file', + appendRule: '- If file exists: Append with `## Iteration {step_iteration}` section', + }, + ja: { + singleHeading: '**レポート出力:** `Report File` に出力してください。', + multiHeading: '**レポート出力:** Report Files に出力してください。', + createRule: '- ファイルが存在しない場合: 新規作成', + appendRule: '- ファイルが存在する場合: `## Iteration {step_iteration}` セクションを追記', + }, +} as const; + +/** + * Generate report output instructions from step.report config. + * Returns undefined if step has no report or no reportDir. + * + * This replaces the manual `order:` fields and instruction_template + * report output blocks that were previously hand-written in each YAML. + */ +function renderReportOutputInstruction( step: WorkflowStep, - context: InstructionContext + context: InstructionContext, + language: Language, +): string | undefined { + if (!step.report || !context.reportDir) return undefined; + + const s = REPORT_OUTPUT_STRINGS[language]; + const isMulti = Array.isArray(step.report); + const heading = isMulti ? s.multiHeading : s.singleHeading; + const appendRule = s.appendRule.replace('{step_iteration}', String(context.stepIteration)); + + return [heading, s.createRule, appendRule].join('\n'); +} + +/** + * Render the Workflow Context section. + */ +function renderWorkflowContext( + step: WorkflowStep, + context: InstructionContext, + language: Language, ): string { - let instruction = step.instructionTemplate; + const s = SECTION_STRINGS[language]; + const lines: string[] = [ + s.workflowContext, + `- ${s.iteration}: ${context.iteration}/${context.maxIterations}${s.iterationWorkflowWide}`, + `- ${s.stepIteration}: ${context.stepIteration}${s.stepIterationTimes}`, + `- ${s.step}: ${step.name}`, + ]; + + // Report info (only if step has report config AND reportDir is available) + if (step.report && context.reportDir) { + lines.push(`- ${s.reportDirectory}: ${context.reportDir}/`); + + if (typeof step.report === 'string') { + // Single file (string form) + lines.push(`- ${s.reportFile}: ${context.reportDir}/${step.report}`); + } else if (isReportObjectConfig(step.report)) { + // Object form (name + order + format) + lines.push(`- ${s.reportFile}: ${context.reportDir}/${step.report.name}`); + } else { + // Multiple files (ReportConfig[] form) + lines.push(`- ${s.reportFiles}:`); + for (const file of step.report as ReportConfig[]) { + lines.push(` - ${file.label}: ${context.reportDir}/${file.path}`); + } + } + } + + return lines.join('\n'); +} + +/** + * Replace template placeholders in the instruction_template body. + * + * These placeholders may still be used in instruction_template for + * backward compatibility or special cases. + */ +function replaceTemplatePlaceholders( + template: string, + step: WorkflowStep, + context: InstructionContext, +): string { + let result = template; + + // These placeholders are also covered by auto-injected sections + // (User Request, Previous Response, Additional User Inputs), but kept here + // for backward compatibility with workflows that still embed them in + // instruction_template (e.g., research.yaml, magi.yaml). + // New workflows should NOT use {task} or {user_inputs} in instruction_template + // since they are auto-injected as separate sections. // Replace {task} - instruction = instruction.replace(/\{task\}/g, escapeTemplateChars(context.task)); + result = result.replace(/\{task\}/g, escapeTemplateChars(context.task)); // Replace {iteration}, {max_iterations}, and {step_iteration} - instruction = instruction.replace(/\{iteration\}/g, String(context.iteration)); - instruction = instruction.replace(/\{max_iterations\}/g, String(context.maxIterations)); - instruction = instruction.replace(/\{step_iteration\}/g, String(context.stepIteration)); + result = result.replace(/\{iteration\}/g, String(context.iteration)); + result = result.replace(/\{max_iterations\}/g, String(context.maxIterations)); + result = result.replace(/\{step_iteration\}/g, String(context.stepIteration)); // Replace {previous_response} if (step.passPreviousResponse) { if (context.previousOutput) { - instruction = instruction.replace( + result = result.replace( /\{previous_response\}/g, - escapeTemplateChars(context.previousOutput.content) + escapeTemplateChars(context.previousOutput.content), ); } else { - instruction = instruction.replace(/\{previous_response\}/g, ''); + result = result.replace(/\{previous_response\}/g, ''); } } - // Replace {git_diff} - const gitDiff = getGitDiff(context.cwd); - instruction = instruction.replace(/\{git_diff\}/g, gitDiff); - // Replace {user_inputs} const userInputsStr = context.userInputs.join('\n'); - instruction = instruction.replace( + result = result.replace( /\{user_inputs\}/g, - escapeTemplateChars(userInputsStr) + escapeTemplateChars(userInputsStr), ); - // Replace {report_dir} with the directory name, keeping paths relative. - // In worktree mode, a symlink from cwd/.takt/reports → projectCwd/.takt/reports - // ensures the relative path resolves correctly without embedding absolute paths - // that could cause agents to operate on the wrong repository. + // Replace {report_dir} if (context.reportDir) { - instruction = instruction.replace(/\{report_dir\}/g, context.reportDir); + result = result.replace(/\{report_dir\}/g, context.reportDir); } - // Append status_rules_prompt with localized header if present - if (step.statusRulesPrompt) { - const statusHeader = renderStatusRulesHeader(context.language ?? 'en'); - instruction = `${instruction}\n\n${statusHeader}\n${step.statusRulesPrompt}`; + // Replace {report:filename} with reportDir/filename + if (context.reportDir) { + result = result.replace(/\{report:([^}]+)\}/g, (_match, filename: string) => { + return `${context.reportDir}/${filename}`; + }); } - // Prepend execution context metadata so agents see it first. - // Now language-aware, so no need to hide it at the end. - const metadata = buildExecutionMetadata(context); - instruction = `${renderExecutionMetadata(metadata)}\n${instruction}`; - - return instruction; + return result; +} + +/** + * Build instruction from template with context values. + * + * Generates a complete instruction by auto-injecting standard sections + * around the step-specific instruction_template content: + * + * 1. Execution Context (working directory, rules) — always + * 2. Workflow Context (iteration, step, report info) — always + * 3. User Request ({task}) — unless template contains {task} + * 4. Previous Response — if passPreviousResponse and has content, unless template contains {previous_response} + * 5. Additional User Inputs — unless template contains {user_inputs} + * 6. Instructions header + instruction_template content — always + * 7. Status Output Rules — if rules exist + * + * Template placeholders ({task}, {previous_response}, etc.) are still replaced + * within the instruction_template body for backward compatibility. + * When a placeholder is present in the template, the corresponding + * auto-injected section is skipped to avoid duplication. + */ +export function buildInstruction( + step: WorkflowStep, + context: InstructionContext, +): string { + const language = context.language ?? 'en'; + const s = SECTION_STRINGS[language]; + const sections: string[] = []; + + // 1. Execution context metadata (working directory + rules + edit permission) + const metadata = buildExecutionMetadata(context, step.edit); + sections.push(renderExecutionMetadata(metadata)); + + // 2. Workflow Context (iteration, step, report info) + sections.push(renderWorkflowContext(step, context, language)); + + // Skip auto-injection for sections whose placeholders exist in the template, + // to avoid duplicate content. Templates using placeholders handle their own layout. + const tmpl = step.instructionTemplate; + const hasTaskPlaceholder = tmpl.includes('{task}'); + const hasPreviousResponsePlaceholder = tmpl.includes('{previous_response}'); + const hasUserInputsPlaceholder = tmpl.includes('{user_inputs}'); + + // 3. User Request (skip if template embeds {task} directly) + if (!hasTaskPlaceholder) { + sections.push(`${s.userRequest}\n${escapeTemplateChars(context.task)}`); + } + + // 4. Previous Response (skip if template embeds {previous_response} directly) + if (step.passPreviousResponse && context.previousOutput && !hasPreviousResponsePlaceholder) { + sections.push( + `${s.previousResponse}\n${escapeTemplateChars(context.previousOutput.content)}`, + ); + } + + // 5. Additional User Inputs (skip if template embeds {user_inputs} directly) + if (!hasUserInputsPlaceholder) { + const userInputsStr = context.userInputs.join('\n'); + sections.push(`${s.additionalUserInputs}\n${escapeTemplateChars(userInputsStr)}`); + } + + // 6a. Report output instruction (auto-generated from step.report) + // If ReportObjectConfig has an explicit `order:`, use that (backward compat). + // Otherwise, auto-generate from the report declaration. + if (step.report && isReportObjectConfig(step.report) && step.report.order) { + const processedOrder = replaceTemplatePlaceholders(step.report.order.trimEnd(), step, context); + sections.push(processedOrder); + } else { + const reportInstruction = renderReportOutputInstruction(step, context, language); + if (reportInstruction) { + sections.push(reportInstruction); + } + } + + // 6b. Instructions header + instruction_template content + const processedTemplate = replaceTemplatePlaceholders( + step.instructionTemplate, + step, + context, + ); + sections.push(`${s.instructions}\n${processedTemplate}`); + + // 6c. Report format (appended after instruction_template, from ReportObjectConfig) + if (step.report && isReportObjectConfig(step.report) && step.report.format) { + const processedFormat = replaceTemplatePlaceholders(step.report.format.trimEnd(), step, context); + sections.push(processedFormat); + } + + // 7. Status rules (auto-generated from rules) + if (step.rules && step.rules.length > 0) { + const statusHeader = renderStatusRulesHeader(language); + const generatedPrompt = generateStatusRulesFromRules(step.name, step.rules, language); + sections.push(`${statusHeader}\n${generatedPrompt}`); + } + + return sections.join('\n\n'); } diff --git a/src/workflow/transitions.ts b/src/workflow/transitions.ts index 5aff143..5f2395a 100644 --- a/src/workflow/transitions.ts +++ b/src/workflow/transitions.ts @@ -1,116 +1,26 @@ /** * Workflow state transition logic * - * Handles determining the next step based on agent response status - * and transition conditions defined in the workflow configuration. + * Handles determining the next step based on rules-based routing. */ import type { - WorkflowConfig, WorkflowStep, - Status, - TransitionCondition, } from '../models/types.js'; -import { COMPLETE_STEP, ABORT_STEP } from './constants.js'; /** - * Check if status matches transition condition. + * Determine next step using rules-based detection. + * Returns the next step name from the matched rule, or null if no rule matched. */ -export function matchesCondition( - status: Status, - condition: TransitionCondition -): boolean { - if (condition === 'always') { - return true; - } - - // Map status to condition - const statusConditionMap: Record = { - done: ['done'], - blocked: ['blocked'], - approved: ['approved'], - rejected: ['rejected'], - improve: ['improve'], - answer: ['answer'], - pending: [], - in_progress: [], - cancelled: [], - interrupted: [], // Interrupted is handled separately - }; - - const matchingConditions = statusConditionMap[status] || []; - return matchingConditions.includes(condition); -} - -/** - * Handle case when no status marker is found in agent output. - */ -export function handleNoStatus( +export function determineNextStepByRules( step: WorkflowStep, - config: WorkflowConfig -): string { - const behavior = step.onNoStatus || 'complete'; - - switch (behavior) { - case 'stay': - // Stay on current step (original behavior, may cause loops) - return step.name; - - case 'continue': { - // Try to find done/always transition, otherwise find next step in workflow - for (const transition of step.transitions) { - if (transition.condition === 'done' || transition.condition === 'always') { - return transition.nextStep; - } - } - // Find next step in workflow order - const stepIndex = config.steps.findIndex(s => s.name === step.name); - const nextStep = config.steps[stepIndex + 1]; - if (stepIndex >= 0 && nextStep) { - return nextStep.name; - } - return COMPLETE_STEP; - } - - case 'complete': - default: - // Try to find done/always transition, otherwise complete workflow - for (const transition of step.transitions) { - if (transition.condition === 'done' || transition.condition === 'always') { - return transition.nextStep; - } - } - return COMPLETE_STEP; + ruleIndex: number, +): string | null { + const rule = step.rules?.[ruleIndex]; + if (!rule) { + return null; } -} - -/** - * Determine next step based on current status. - */ -export function determineNextStep( - step: WorkflowStep, - status: Status, - config: WorkflowConfig -): string { - // If interrupted, abort immediately - if (status === 'interrupted') { - return ABORT_STEP; - } - - // Check transitions in order - for (const transition of step.transitions) { - if (matchesCondition(status, transition.condition)) { - return transition.nextStep; - } - } - - // If status is 'in_progress' (no status marker found), use onNoStatus behavior - if (status === 'in_progress') { - return handleNoStatus(step, config); - } - - // Unexpected status - treat as done and complete - return COMPLETE_STEP; + return rule.next; } /** diff --git a/src/workflow/types.ts b/src/workflow/types.ts index bfabea8..d41f82e 100644 --- a/src/workflow/types.ts +++ b/src/workflow/types.ts @@ -12,7 +12,8 @@ import type { PermissionHandler, AskUserQuestionHandler } from '../claude/proces /** Events emitted by workflow engine */ export interface WorkflowEvents { 'step:start': (step: WorkflowStep, iteration: number) => void; - 'step:complete': (step: WorkflowStep, response: AgentResponse) => void; + 'step:complete': (step: WorkflowStep, response: AgentResponse, instruction: string) => void; + 'step:report': (step: WorkflowStep, filePath: string, fileName: string) => void; 'step:blocked': (step: WorkflowStep, response: AgentResponse) => void; 'step:user_input': (step: WorkflowStep, userInput: string) => void; 'workflow:complete': (state: WorkflowState) => void;