--- name: bugfix-executor-generator description: This skill should be used when the user asks to 'generate bugfix executor', 'create dynamic executor', 'prepare bugfix execution skill', or needs to generate a self-contained executor skill with architect review workflow. allowed-tools: Read, Write, Edit, Bash, Glob, Grep version: 1.1 --- # Bugfix Executor Generator v1.1 Generate dynamic executor skill với Hybrid approach. --- ## WHAT'S NEW IN v1.1 - **Hybrid Approach**: Architect Design → Implement → Architect Review - **Scope Detection**: Tự động detect Backend\/Frontend\/Mobile từ files - **Architect Review**: Gọi architect agents review code sau implement - **Multi-Round Fix**: Fix issues từ architect review --- ## WHAT THIS DOES 1. Đọc PLAN.md từ `.claude\/plans\/fix-{ISSUE_ID}\/` 2. Detect scope từ files (Backend\/Frontend\/Mobile) 3. Tạo skill mới tại `.claude\/skills\/_dynamic\/exec-fix-{ISSUE_ID}\/SKILL.md` 4. Skill mới chứa: - Toàn bộ context của task - Architect Review sau implement - Multi-round fix nếu có issues 5. Khi execute xong, skill tự xóa chính nó --- ## USAGE ```bash \/bugfix-executor-generator # Find plan and generate skill \/bugfix-executor-generator .claude\/plans\/fix-123\/ # Specific plan ``` --- ## GENERATED SKILL LOCATION ``` .claude\/skills\/_dynamic\/ ├── exec-fix-123\/ │ ├── SKILL.md # Generated executor for Issue #123 │ └── context.json # Embedded context from plan ├── exec-fix-456\/ │ └── SKILL.md # Generated executor for Issue #456 └── ... ``` --- ## WORKFLOW ### Step 1: Parse Plan & Detect Scope ```typescript \/\/ Find plan PLAN_DIR = ".claude\/plans\/fix-${ISSUE_ID}" PLAN_FILE = "${PLAN_DIR}\/PLAN.md" ISSUE_ID = extractIssueId(PLAN_DIR) \/\/ Read plan ${PLAN_FILE}<\/Read> \/\/ Parse to extract structured data plan = { issue_id: ISSUE_ID, summary: extractSection("Summary"), root_cause: extractSection("Root Cause Analysis"), implementation_steps: extractImplementationSteps(), files_to_change: extractFilesTable(), test_cases: extractTestCases() } \/\/ DETECT SCOPE from files scope = [] FOR file IN plan.files_to_change: IF file.path.includes("packages\/backend") || file.path.includes(".entity.ts") || file.path.includes(".service.ts") || file.path.includes(".controller.ts"): scope.push("Backend") IF file.path.includes("packages\/frontend") || file.path.includes(".vue") || file.path.includes("components\/"): scope.push("Frontend") IF file.path.includes("packages\/mobile") || file.path.includes(".dart"): scope.push("Mobile") scope = [...new Set(scope)] \/\/ Remove duplicates IF scope.length == 0: scope = ["Backend"] \/\/ Default plan.scope = scope \/\/ Load detail files IF exists("${PLAN_DIR}\/details\/data-flow.md"): ${PLAN_DIR}\/details\/data-flow.md<\/Read> plan.data_flow = content IF exists("${PLAN_DIR}\/details\/impact-analysis.md"): ${PLAN_DIR}\/details\/impact-analysis.md<\/Read> plan.impact = content IF exists("${PLAN_DIR}\/details\/test-cases.md"): ${PLAN_DIR}\/details\/test-cases.md<\/Read> plan.test_cases_detail = content ``` ### Step 2: Generate Dynamic Skill with Hybrid Approach ```typescript \/\/ Create skill directory SKILL_DIR = ".claude\/skills\/_dynamic\/exec-fix-${ISSUE_ID}" mkdir -p ${SKILL_DIR} \/\/ Generate SKILL.md with embedded context + Hybrid approach skill_content = ` --- name: exec-fix-${ISSUE_ID} description: Execute bugfix #${ISSUE_ID} with Hybrid approach. Auto-generated, will self-delete after completion. allowed-tools: Read, Write, Edit, Bash, Glob, Grep, Task, TodoWrite version: 1.0 auto-generated: true issue-id: ${ISSUE_ID} scope: ${plan.scope.join(", ")} hooks: PostToolUse: - matcher: { tool_name: "Edit" } command: "bash ~\/.claude\/scripts\/auto-lint.sh" - matcher: { tool_name: "Write" } command: "bash ~\/.claude\/scripts\/auto-lint.sh" --- # Executor for Bugfix #${ISSUE_ID} > **Auto-generated skill** - Hybrid Approach: Implement → Architect Review → Fix > **Scope:** ${plan.scope.join(", ")} --- ## EMBEDDED CONTEXT ### Issue Summary ${plan.summary} ### Root Cause ${plan.root_cause} ### Scope ${plan.scope.join(", ")} ### Implementation Steps ${plan.implementation_steps.map((step, i) => ` #### Step ${i + 1}: ${step.title} - **File:** ${step.file} - **Sub-steps:** ${step.sub_steps.map(s => ` - ${s.description}`).join('\n')} ${step.logic ? `- **Logic:**\n\`\`\`\n${step.logic}\n\`\`\`` : ''} `).join('\n')} ### Files to Change | # | File | Change | Lines | |---|------|--------|-------| ${plan.files_to_change.map((f, i) => `| ${i + 1} | \`${f.file}\` | ${f.change} | ~${f.lines} |`).join('\n')} ${plan.data_flow ? ` ### Data Flow ${plan.data_flow} ` : ''} --- ## EXECUTION (Hybrid Approach) ### Phase 1: Initialize \`\`\`typescript ISSUE_ID = "${ISSUE_ID}" PLAN_DIR = ".claude\/plans\/fix-${ISSUE_ID}" SKILL_DIR = ".claude\/skills\/_dynamic\/exec-fix-${ISSUE_ID}" PROGRESS_FILE = "\${PLAN_DIR}\/.progress-${ISSUE_ID}.json" SCOPE = ${JSON.stringify(plan.scope)} \/\/ Load or create progress IF exists(PROGRESS_FILE): progress = JSON.parse(\${PROGRESS_FILE}<\/Read>) current_step = progress.current_step ELSE: current_step = 0 progress = { issue_id: "${ISSUE_ID}", scope: SCOPE, steps_completed: [], current_step: 0, architect_review: null, review_rounds: 0 } console.log("🚀 Executing Bugfix #${ISSUE_ID} (Hybrid Approach)") console.log("📋 Steps: ${plan.implementation_steps.length}") console.log("📁 Scope: ${plan.scope.join(', ')}") \`\`\` ### Phase 2: Implement Steps ${plan.implementation_steps.map((step, i) => ` #### Execute Step ${i + 1}: ${step.title} \`\`\`typescript IF current_step <= ${i}: console.log("\\n${"=".repeat(50)}") console.log("STEP ${i + 1}: ${step.title}") console.log("${"=".repeat(50)}\\n") \/\/ 1. Read target file ${step.file}<\/Read> \/\/ 2. Implement changes ${step.sub_steps.map(s => `\/\/ - ${s.description}`).join('\n ')} \/\/ 3. Verify tsc_result = npx tsc --noEmit 2>&1 | head -20<\/Bash> IF tsc_result.hasErrors: fixErrors() lint_result = npx eslint ${step.file} --quiet 2>&1<\/Bash> IF lint_result.hasErrors: npx eslint ${step.file} --fix<\/Bash> \/\/ 4. Save progress progress.current_step = ${i + 1} progress.steps_completed.push(${i}) \${JSON.stringify(progress, null, 2)}<\/Write> console.log("✅ Step ${i + 1} COMPLETE") \`\`\` `).join('\n')} ### Phase 3: Architect Review (Hybrid) \`\`\`typescript console.log("\\n🏗️ ARCHITECT REVIEW") console.log("Scope: ${plan.scope.join(', ')}") \/\/ Get list of modified files modified_files = ${JSON.stringify(plan.files_to_change.map(f => f.file))} \/\/ Run architect review based on scope (PARALLEL) ${plan.scope.includes('Backend') ? ` ## Bugfix Code Review Request **Issue:** #${ISSUE_ID} **Summary:** ${plan.summary} **Modified Files:** ${plan.files_to_change.filter(f => f.file.includes('packages\/backend') || f.file.includes('.entity.ts') || f.file.includes('.service.ts') || f.file.includes('.controller.ts') ).map(f => `- ${f.file}`).join('\n')} **Review Focus:** 1. Code correctness - Does the fix address the root cause? 2. NestJS best practices - Proper DI, decorators, error handling 3. Performance - No N+1 queries, proper caching 4. Security - No vulnerabilities introduced 5. Side effects - Any unintended changes? **Output Format:** - PASS: Code is good - ISSUES: List specific issues with file:line and fix suggestion <\/Task> ` : '\/\/ Skip Backend review - not in scope'} ${plan.scope.includes('Frontend') ? ` ## Bugfix Code Review Request **Issue:** #${ISSUE_ID} **Summary:** ${plan.summary} **Modified Files:** ${plan.files_to_change.filter(f => f.file.includes('packages\/frontend') || f.file.includes('.vue') ).map(f => `- ${f.file}`).join('\n')} **Review Focus:** 1. Vue 3 best practices - Composition API, reactivity 2. Component structure - Props, emits, slots 3. State management - Pinia store usage 4. Performance - Computed vs methods, v-memo 5. Accessibility - ARIA, keyboard navigation **Output Format:** - PASS: Code is good - ISSUES: List specific issues with file:line and fix suggestion <\/Task> ` : '\/\/ Skip Frontend review - not in scope'} ${plan.scope.includes('Mobile') ? ` ## Bugfix Code Review Request **Issue:** #${ISSUE_ID} **Summary:** ${plan.summary} **Modified Files:** ${plan.files_to_change.filter(f => f.file.includes('packages\/mobile') || f.file.includes('.dart') ).map(f => `- ${f.file}`).join('\n')} **Review Focus:** 1. Flutter best practices - Widget composition 2. State management - BLoC\/Provider patterns 3. Performance - Build method optimization 4. Platform handling - iOS\/Android differences **Output Format:** - PASS: Code is good - ISSUES: List specific issues with file:line and fix suggestion <\/Task> ` : '\/\/ Skip Mobile review - not in scope'} \/\/ Collect review results architect_reviews = await collectReviewResults() progress.architect_review = architect_reviews \${JSON.stringify(progress, null, 2)}<\/Write> \`\`\` ### Phase 4: Fix Issues from Review (if any) \`\`\`typescript console.log("\\n🔧 FIX REVIEW ISSUES") IF architect_reviews.hasIssues: progress.review_rounds += 1 IF progress.review_rounds > 2: console.log("⚠️ Max review rounds (2) reached. Manual review needed.") ELSE: FOR issue IN architect_reviews.issues: console.log(\`Fixing: \${issue.file}:\${issue.line}\`) \/\/ Apply fix \${issue.file}<\/Read> \/\/ Edit to fix the issue Apply fix: \${issue.suggestion}<\/Edit> \/\/ Re-run verification npx tsc --noEmit 2>&1 | head -20<\/Bash> npm run lint -- --quiet 2>&1 | head -20<\/Bash> \/\/ Update progress \${JSON.stringify(progress, null, 2)}<\/Write> console.log("✅ Review issues fixed") ELSE: console.log("✅ No issues from architect review") \`\`\` ### Phase 5: Final Verification \`\`\`typescript console.log("\\n🔍 FINAL VERIFICATION") npx tsc --noEmit<\/Bash> npm run lint -- --quiet<\/Bash> npm run test -- --passWithNoTests<\/Bash> await allBackgroundTasks() IF all_passed: console.log("✅ All verifications passed!") ELSE: console.log("❌ Some verifications failed") showFailures() \`\`\` ### Phase 6: Self-Cleanup \`\`\`typescript console.log("\\n🧹 CLEANUP") IF all_passed: ASK USER: "Delete executor skill and temp files? (Y\/N)" IF user_confirms: \/\/ Delete progress file rm "\${PROGRESS_FILE}" \/\/ Delete detail files rm -rf "\${PLAN_DIR}\/details\/" \/\/ DELETE THIS SKILL (self-cleanup) rm -rf "\${SKILL_DIR}" console.log("🎉 Bugfix #${ISSUE_ID} complete! Skill self-deleted.") console.log("ℹ To cleanup plan: rm -rf \${PLAN_DIR}") \`\`\` --- ## STRICT RULES 1. **NO SKIP** - Execute each step in order 2. **VERIFY** - TypeScript + Lint after each step 3. **ARCHITECT REVIEW** - Run after all steps complete 4. **FIX ISSUES** - Apply fixes from architect review (max 2 rounds) 5. **SAVE** - Progress after each phase 6. **SELF-DELETE** - Remove this skill after completion ` ${skill_content} <\/Write> \/\/ Save context as JSON for reference ${JSON.stringify(plan, null, 2)} <\/Write> console.log(` ✅ Dynamic executor skill created (Hybrid Approach)! Location: ${SKILL_DIR}\/SKILL.md Issue: #${ISSUE_ID} Scope: ${plan.scope.join(", ")} Steps: ${plan.implementation_steps.length} Flow: 1. Implement all steps 2. Architect Review (${plan.scope.join(" + ")}) 3. Fix issues (if any) 4. Final verification 5. Self-cleanup To execute: \/exec-fix-${ISSUE_ID} This skill will self-delete after successful completion. `) ``` --- ## CLEANUP Skill tự xóa sau khi hoàn thành, hoặc manual: ```bash # Delete specific dynamic skill rm -rf .claude\/skills\/_dynamic\/exec-fix-123\/ # Delete all dynamic skills rm -rf .claude\/skills\/_dynamic\/ ```