@haposoft/cafekit 0.3.2 → 0.3.6

package/src/antigravity/workflows/impact-analysis.md

@@ -0,0 +1,735 @@
+---
+description: Phân tích tác động của code changes lên tính năng và hành động người dùng
+allowed-tools: Read, Glob, Grep, Bash, WebSearch
+argument-hint: <feature-name> [--from <branch>]
+---
+
+# /impact-analysis - Phân Tích Tác Động Tính Năng
+
+Sử dụng workflow này sau khi hoàn thành feature (sau `/code`) hoặc trước khi commit.
+
+<background_information>
+- **Mission**: Comprehensive impact analysis for completed features to prevent regression bugs
+- **Success Criteria**:
+  - All affected features identified with user actions mapped
+  - Edge cases detected and documented
+  - Test scenarios generated for all critical paths
+  - Regression risks assessed with mitigation recommendations
+  - Cross-referenced with spec artifacts when available
+- **Timing**: After implementing all tasks for a feature, before running `/test`
+- **Helper Scripts**: Optionally use advanced analysis scripts from `.claude/skills/impact-analysis/scripts/` for automated dependency analysis, AST-based semantic analysis, and risk scoring
+</background_information>
+
+<instructions>
+## Core Task
+Analyze the impact of code changes for feature **$ARGUMENTS** to identify affected features, user actions, edge cases, and generate comprehensive test scenarios.
+
+## Helper Scripts (Optional Enhancement)
+
+If available in `.claude/skills/impact-analysis/scripts/`, these scripts provide advanced automated analysis:
+
+**1. Dependency Analysis**:
+```bash
+.claude/skills/impact-analysis/scripts/find-dependencies.sh <file>
+```
+- Finds all files importing/using the changed file
+- Identifies API consumers and component usage
+- Provides impact recommendations
+
+**2. AST-Based Analysis** (for TypeScript/JavaScript):
+```bash
+node .claude/skills/impact-analysis/scripts/ast-analyze.js <file>
+```
+- Detects function signature changes
+- Identifies breaking changes
+- Compares before/after versions
+
+**3. Risk Calculation**:
+```bash
+node .claude/skills/impact-analysis/scripts/calculate-risk.js
+```
+- Calculates risk score (0-25)
+- Determines risk level (CRITICAL/HIGH/MEDIUM/LOW)
+- Provides actionable recommendations with time estimates
+
+**4. Complete Analysis**:
+```bash
+.claude/skills/impact-analysis/scripts/run-analysis.sh
+```
+- Runs all techniques
+- Generates comprehensive markdown report
+
+**Usage Strategy**:
+- Try to use helper scripts first for automated analysis
+- Fall back to manual grep/analysis if scripts not available
+- Combine automated results with manual insights for best coverage
+
+## Execution Steps
+
+### Step 0: Validate Context & Determine Scope
+
+**Parse Arguments**:
+- Extract feature name from `$ARGUMENTS` (first argument)
+- Check for flags: `--from <branch>` for branch-based analysis
+- If no arguments: Analyze uncommitted changes
+
+**Load Spec Context (if feature name provided)**:
+- Check if `.specs/$FEATURE_NAME/` exists
+- If exists:
+  - Read `.specs/$FEATURE_NAME/spec.json` for metadata
+  - Read `.specs/$FEATURE_NAME/requirements.md` for requirements
+  - Read `.specs/$FEATURE_NAME/design.md` for design decisions
+  - Read `.specs/$FEATURE_NAME/tasks.md` for task list
+  - Note: This enables cross-referencing and coverage analysis
+- If not exists: Continue with standalone analysis (no spec context)
+
+**Determine Analysis Scope**:
+- **Feature-based** (feature name provided): Analyze all changes since feature start
+  - Use git log to find first commit mentioning feature
+  - Or analyze all uncommitted + recent commits
+- **Branch-based** (`--from` flag): Compare current branch with specified branch
+  - Example: `--from main` analyzes all changes since branching from main
+- **Uncommitted** (no args): Analyze only uncommitted changes
+  - Use `git diff HEAD` for staged/unstaged changes
+
+### Step 1: Detect & Classify Changes
+
+**Execute Git Analysis**:
+```bash
+# Based on scope from Step 0:
+# Feature-based: git diff <feature-start-commit>..HEAD
+# Branch-based: git diff <branch>...HEAD
+# Uncommitted: git diff HEAD
+```
+
+**Classify Changed Files**:
+- **Backend**: Files in `src/api/`, `src/services/`, `src/lib/`, `src/utils/`
+- **Frontend**: Files in `src/components/`, `src/pages/`, `src/app/`, `src/hooks/`
+- **Database**: Files in `prisma/`, `migrations/`, `schema/`, `*.sql`
+- **Config**: Files like `package.json`, `tsconfig.json`, `.env`, config files
+- **Tests**: Files in `__tests__/`, `*.test.*`, `*.spec.*`
+
+**Count & Summarize**:
+- Total files changed
+- Files per category
+- Lines added/removed
+- Commits involved (if applicable)
+
+### Step 2: Map Features & User Actions
+
+**Pattern-Based Feature Detection**:
+
+Use file path patterns to identify affected features:
+
+| File Pattern | Feature | User Actions |
+|-------------|---------|--------------|
+| `**/api/auth**`, `**/auth**` | Authentication & Login | Login, Logout, Sign Up, Password Reset, Session Management, 2FA |
+| `**/api/users**`, `**/user**`, `**/profile**` | User Profile Management | View Profile, Edit Profile, Upload Avatar, Change Settings, Update Bio |
+| `**/api/posts**`, `**/post**`, `**/article**` | Post Management | Create Post, Edit Post, Delete Post, View Posts, Like Post, Share Post |
+| `**/api/comments**`, `**/comment**`, `**/reply**` | Comment System | Add Comment, Edit Comment, Delete Comment, View Comments, Reply |
+| `**/api/payments**`, `**/payment**`, `**/checkout**` | Payment Processing | Make Payment, Process Refund, View History, Update Payment Method |
+| `**/api/notifications**`, `**/notification**` | Notification System | Send Notification, View Notifications, Mark as Read, Settings |
+| `**/components/auth**` | Login/Signup UI | Display Forms, Handle Validation, Show Errors |
+| `**/hooks/useAuth**` | Authentication State | Check Login Status, Get User Info, Handle Logout |
+| `**/prisma/schema**`, `**/migrations**` | Database Structure | All Database Operations, Queries, Mutations |
+| `**/middleware/auth**` | Access Control | Check Permissions, Protect Routes, Validate Tokens |
+
+**Content-Based Feature Detection**:
+
+Scan file contents for keywords to detect additional features:
+
+| Keywords | Feature | User Actions |
+|----------|---------|--------------|
+| `login`, `signin`, `authenticate` | Authentication & Login | Login, Logout, Session Management |
+| `profile`, `avatar`, `bio` | User Profile Management | View Profile, Edit Profile, Upload Avatar |
+| `post`, `article`, `content` | Post Management | Create, Edit, Delete, View Posts |
+| `comment`, `reply` | Comment System | Add, Edit, Delete Comments |
+| `payment`, `checkout`, `stripe`, `paypal` | Payment Processing | Make Payment, Process Refund |
+| `search`, `filter`, `query` | Search & Filter | Search Content, Apply Filters, Sort Results |
+| `upload`, `file`, `image` | File Upload | Upload File, Delete File, View Files |
+| `email`, `send mail` | Email System | Send Email, Email Verification |
+
+**Generate Feature Impact Map**:
+- List all detected features
+- Map user actions to each feature
+- Create impact scenarios: "When user tries to [action], [potential impact]"
+
+### Step 3: Find Dependencies & Affected Files
+
+**Automated Dependency Analysis** (using helper script):
+
+If helper script available, run comprehensive analysis:
+```bash
+# Run dependency analysis script for each changed file
+.claude/skills/impact-analysis/scripts/find-dependencies.sh <file>
+```
+
+**Manual Dependency Analysis** (fallback):
+
+For each changed file, find:
+1. **Direct imports**: Files that import the changed file
+   ```bash
+   grep -r "from.*<filename>" src/
+   grep -r "import.*<filename>" src/
+   ```
+
+2. **Function/Class usage**: Files that call functions or use classes from changed file
+   ```bash
+   grep -r "<function-name>\|<class-name>" src/
+   ```
+
+3. **API consumers**: If API changed, find frontend components calling it
+   ```bash
+   grep -r "fetch.*<api-endpoint>" src/
+   grep -r "axios.*<api-endpoint>" src/
+   ```
+
+4. **Component usage**: If component changed, find parent components
+   ```bash
+   grep -r "<ComponentName>" src/
+   ```
+
+**Count Affected Files**:
+- Total affected files
+- Breakdown by category (backend/frontend/tests)
+- Highlight high-impact files (used in many places)
+
+### Step 4: Identify Edge Cases
+
+**Automated Edge Case Detection**:
+
+**Option 1: Use Helper Scripts** (if available):
+```bash
+# Run AST analysis for TypeScript/JavaScript files
+node .claude/skills/impact-analysis/scripts/ast-analyze.js <file>
+
+# Run risk calculation
+node .claude/skills/impact-analysis/scripts/calculate-risk.js
+```
+
+**Option 2: Manual Detection** (fallback):
+
+Scan changed files for common issues:
+
+1. **Null/Undefined Handling**:
+   ```bash
+   # Find property access without optional chaining
+   grep -E "\.[a-zA-Z]+" <file> | grep -v "?\."
+   ```
+   - Issue: Potential null pointer errors
+   - Action: Add null checks or optional chaining
+
+2. **Error Handling**:
+   ```bash
+   # Find async calls without try-catch
+   grep -E "await|fetch|axios" <file>
+   grep -E "try\s*\{" <file>
+   ```
+   - Issue: Missing error handling
+   - Action: Add try-catch blocks
+
+3. **Boundary Conditions**:
+   - Array access without length check
+   - Division without zero check
+   - String operations without empty check
+
+4. **Race Conditions**:
+   - Multiple async operations without proper sequencing
+   - State updates without locks/mutexes
+   - Concurrent API calls
+
+**Categorize by Severity**:
+- **Critical**: Will cause crashes or data loss
+- **Important**: Will cause incorrect behavior
+- **Minor**: Edge cases that rarely occur
+
+### Step 5: Cross-Reference with Spec (if available)
+
+**If spec context loaded in Step 0**:
+
+**Requirements Coverage**:
+- Map changed files to requirement IDs
+- Check if all requirements are implemented
+- Identify partially implemented requirements
+- Flag missing implementations
+
+**Design Compliance**:
+- Verify API endpoints match design
+- Check database schema matches design
+- Validate component structure follows design
+- Note any deviations
+
+**Task Completion**:
+- Check task status in `tasks.md`
+- Count completed vs pending tasks
+- Identify blocked tasks
+- Calculate completion percentage
+
+**Output**:
+```markdown
+## Requirements Coverage
+- FR-1: User Login ✅ Implemented
+- FR-2: Password Reset ✅ Implemented
+- FR-3: 2FA ⚠️ Partially implemented
+
+## Design Compliance
+- API endpoints: ✅ Match design
+- Database schema: ✅ Match design
+- Component structure: ⚠️ Deviation detected
+
+## Task Status
+- Completed: 8/10 tasks (80%)
+- Pending: Task #9, Task #10
+```
+
+### Step 6: Generate Test Scenarios
+
+**Scenario Generation Strategy**:
+
+For each affected feature and user action, generate:
+
+1. **Happy Path Scenarios**:
+   ```markdown
+   Given: [precondition]
+   When: [user action]
+   Then: [expected result]
+   And: [additional verification]
+   ```
+
+2. **Error Handling Scenarios**:
+   ```markdown
+   Given: [precondition]
+   When: [user action with invalid input]
+   Then: [error message shown]
+   And: [system remains stable]
+   ```
+
+3. **Edge Case Scenarios**:
+   ```markdown
+   Given: [edge case condition]
+   When: [user action]
+   Then: [graceful handling]
+   ```
+
+4. **Regression Scenarios**:
+   ```markdown
+   Given: [existing functionality]
+   When: [user performs old action]
+   Then: [still works as before]
+   ```
+
+**Prioritize Scenarios**:
+- **Critical**: Must test before deployment
+- **Important**: Should test before PR
+- **Nice to have**: Test if time permits
+
+### Step 7: Assess Regression Risks
+
+**Automated Risk Assessment** (if helper script available):
+
+```bash
+# Run comprehensive risk calculation
+node .claude/skills/impact-analysis/scripts/calculate-risk.js
+
+# Output includes:
+# - Risk score (0-25)
+# - Risk level (CRITICAL/HIGH/MEDIUM/LOW)
+# - Risk breakdown by factors
+# - Actionable recommendations with time estimates
+```
+
+**Manual Risk Assessment** (fallback):
+
+Categorize changes by risk level:
+
+**High Risk**:
+- Database schema changes
+- Authentication/authorization changes
+- Payment processing changes
+- API contract changes (breaking)
+- Core business logic modifications
+
+**Medium Risk**:
+- UI component refactoring
+- New features with existing integrations
+- Configuration changes
+- Dependency updates
+
+**Low Risk**:
+- Documentation updates
+- Code formatting
+- Minor UI tweaks
+- Test additions
+
+**For Each Risk Area**:
+- Identify what could break
+- List affected user workflows
+- Suggest mitigation strategies
+
+### Step 8: Create Comprehensive Report
+
+**Generate Report Structure**:
+
+```markdown
+# Impact Analysis Report - {Feature Name}
+
+**Generated**: {timestamp}
+**Feature**: {feature-name}
+**Branch**: {current-branch}
+**Commits**: {commit-count}
+
+---
+
+## 📋 Executive Summary
+
+- **Files Changed**: {count}
+- **Features Affected**: {count}
+- **User Scenarios**: {count}
+- **Edge Cases**: {count}
+- **Risk Level**: {HIGH/MEDIUM/LOW}
+- **Estimated Test Time**: {estimate}
+
+{If spec available:}
+- **Requirements Coverage**: {percentage}%
+- **Tasks Completed**: {completed}/{total}
+- **Design Compliance**: {status}
+
+---
+
+## 🔍 Change Analysis
+
+### Files Changed ({count})
+
+**Backend** ({count} files):
+- {list of backend files}
+
+**Frontend** ({count} files):
+- {list of frontend files}
+
+**Database** ({count} files):
+- {list of database files}
+
+**Config** ({count} files):
+- {list of config files}
+
+---
+
+## 🎯 Feature Impact Map
+
+### Feature 1: {Feature Name}
+
+**User Actions Affected**:
+- {action 1}
+- {action 2}
+- {action 3}
+
+**Impact Level**: {HIGH/MEDIUM/LOW}
+
+**Details**: {explanation of impact}
+
+### Feature 2: {Feature Name}
+...
+
+---
+
+## 🎬 Impact Scenarios
+
+### Scenario 1: {Action Name}
+- **When user tries to**: {action}
+- **Potential impact**: {description}
+- **Required action**: {test instruction}
+- **Priority**: {CRITICAL/IMPORTANT/MINOR}
+
+### Scenario 2: {Action Name}
+...
+
+---
+
+## 🔗 Dependencies & Affected Files
+
+### Direct Dependencies ({count})
+- {file 1} → {affected by changes in file X}
+- {file 2} → {affected by changes in file Y}
+
+### Integration Points ({count})
+- {component A} calls {API endpoint B}
+- {page X} uses {component Y}
+
+---
+
+## ⚠️ Edge Cases Detected
+
+### Critical ({count})
+1. **{Issue Title}**
+   - **Location**: {file:line}
+   - **Issue**: {description}
+   - **Impact**: {what could go wrong}
+   - **Action**: {how to fix}
+
+### Important ({count})
+...
+
+### Minor ({count})
+...
+
+---
+
+{If spec available:}
+## 📊 Spec Cross-Reference
+
+### Requirements Coverage
+- FR-1: {requirement} ✅ Implemented
+- FR-2: {requirement} ⚠️ Partially implemented
+- FR-3: {requirement} ❌ Not implemented
+
+### Design Compliance
+- API Endpoints: {status}
+- Database Schema: {status}
+- Component Structure: {status}
+
+### Task Completion
+- Completed: {count}/{total} ({percentage}%)
+- Pending: {list}
+- Blocked: {list}
+
+---
+
+## 🧪 Test Scenarios
+
+### Critical Scenarios (Must Test)
+
+#### Scenario 1: {Name}
+```gherkin
+Given: {precondition}
+When: {action}
+Then: {expected result}
+And: {verification}
+```
+
+**Priority**: CRITICAL
+**Estimated Time**: {minutes}
+
+#### Scenario 2: {Name}
+...
+
+### Important Scenarios (Should Test)
+...
+
+### Regression Scenarios (Verify Old Functionality)
+...
+
+---
+
+## 🎯 Regression Risk Assessment
+
+### High Risk Areas
+1. **{Area Name}**
+   - **Why**: {reason}
+   - **Impact**: {what could break}
+   - **Mitigation**: {how to prevent}
+
+### Medium Risk Areas
+...
+
+### Low Risk Areas
+...
+
+---
+
+## 💡 Recommendations
+
+### Before Testing
+- [ ] {recommendation 1}
+- [ ] {recommendation 2}
+- [ ] {recommendation 3}
+
+### Testing Checklist
+- [ ] Test all {count} critical scenarios
+- [ ] Verify {count} edge cases fixed
+- [ ] Run regression tests
+- [ ] Test on multiple devices/browsers
+- [ ] Performance testing (if applicable)
+
+### Before Deployment
+- [ ] Update documentation
+- [ ] Notify stakeholders
+- [ ] Prepare rollback plan
+- [ ] Monitor metrics post-deployment
+
+---
+
+## 🚀 Next Steps
+
+1. **Review this report** - Understand all impacts
+2. **Fix critical edge cases** - Address {count} critical issues
+3. **Run test scenarios** - Execute all critical tests
+4. **Re-run analysis** - Verify fixes with `/impact-analysis {feature-name}`
+5. **Proceed to testing** - Run `/test`
+6. **Code review** - Run `/review`
+7. **Commit & PR** - Create pull request with this report attached
+
+---
+
+**Report saved to**: `.specs/impact-reports/{feature-name}-{timestamp}.md`
+```
+
+**Save Report**:
+- Create `.specs/impact-reports/` directory if not exists
+- Save report as `{feature-name}-{timestamp}.md`
+- Create symlink `latest.md` pointing to this report
+
+</instructions>
+
+## Tool Guidance
+
+**Read Operations**:
+- Load spec context first (if feature name provided)
+- Read all relevant spec files (spec.json, requirements.md, design.md, tasks.md)
+- Use Grep for dependency analysis and keyword detection
+
+**Bash Operations**:
+- Use git commands for change detection
+- Use grep for pattern matching and content scanning
+- Use find for file discovery
+
+**Analysis Strategy**:
+- Start broad (all changes) then narrow down (specific impacts)
+- Cross-reference with spec when available
+- Prioritize by risk and user impact
+
+**Output**:
+- Generate comprehensive markdown report
+- Save to `.specs/impact-reports/`
+- Display summary in terminal
+
+## Output Description
+
+**Terminal Output** (concise summary):
+
+```markdown
+✅ Impact Analysis Complete
+
+**Feature**: {feature-name}
+**Files Changed**: {count}
+**Features Affected**: {count}
+**User Scenarios**: {count}
+**Edge Cases**: {count} ({critical} critical)
+**Risk Level**: {HIGH/MEDIUM/LOW}
+
+{If spec available:}
+**Requirements Coverage**: {percentage}%
+**Tasks Completed**: {completed}/{total}
+
+📄 **Full Report**: `.specs/impact-reports/{feature-name}-{timestamp}.md`
+
+🚀 **Next Steps**:
+1. Review report and fix {count} critical edge cases
+2. Run `/test` to execute test scenarios
+3. Run `/review` for code quality check
+```
+
+**Report File**: Comprehensive markdown report saved to `.specs/impact-reports/`
+
+## Safety & Fallback
+
+### Error Scenarios
+
+**No Changes Detected**:
+- **Message**: "No changes detected. Check git status or specify files."
+- **Suggested Action**: "Ensure you have uncommitted changes or specify feature name"
+
+**Feature Not Found**:
+- **Message**: "Feature '{name}' not found in .specs/"
+- **Suggested Action**: "Continue with standalone analysis (no spec context)"
+- **Proceed**: Analyze changes without spec cross-reference
+
+**Git Not Available**:
+- **Message**: "Git not available. Cannot detect changes."
+- **Suggested Action**: "Ensure git is installed and repository is initialized"
+- **Fallback**: Ask user to specify files manually
+
+**No Spec Context**:
+- **Warning**: "No spec context available. Analysis will be standalone."
+- **Proceed**: Continue without requirements/design/tasks cross-reference
+- **Note**: Report will not include spec-related sections
+
+### Next Phase
+
+**After Impact Analysis**:
+1. **Review Report**: Read full report at `.specs/impact-reports/latest.md`
+2. **Fix Critical Issues**: Address all critical edge cases
+3. **Run Tests**: Execute `/test` with generated scenarios
+4. **Code Review**: Run `/review` for quality check
+5. **Re-analyze** (optional): Run `/impact-analysis` again to verify fixes
+6. **Commit**: Create commit with report attached to PR
+
+**Workflow Integration**:
+```
+/code {feature-name}
+  ↓ (implement all tasks)
+/impact-analysis {feature-name}
+  ↓ (review report, fix issues)
+/test
+  ↓
+/review
+  ↓
+Commit & PR
+```
+
+## Examples
+
+### Example 1: Feature-Based Analysis
+
+```bash
+# After completing user-authentication feature
+/impact-analysis user-authentication
+
+# Output:
+# - Analyzes all changes for the feature
+# - Cross-references with spec
+# - Maps to Authentication & Login feature
+# - Lists 5 user actions affected
+# - Identifies 3 critical edge cases
+# - Generates 12 test scenarios
+# - Report saved to .specs/impact-reports/user-authentication-20260320.md
+```
+
+### Example 2: Uncommitted Changes
+
+```bash
+# Quick analysis of current work
+/impact-analysis
+
+# Output:
+# - Analyzes uncommitted changes only
+# - No spec context (standalone)
+# - Maps to affected features
+# - Identifies edge cases
+# - Generates test scenarios
+```
+
+### Example 3: Branch-Based Analysis
+
+```bash
+# Analyze all changes in feature branch
+/impact-analysis --from main
+
+# Output:
+# - Compares current branch with main
+# - Analyzes all differences
+# - Comprehensive impact report
+```
+
+## Best Practices
+
+1. **Run After Feature Completion**: Use after implementing all tasks, not after each small change
+2. **Review Report Thoroughly**: Don't skip edge cases and test scenarios
+3. **Fix Critical Issues First**: Address all critical edge cases before testing
+4. **Attach to PR**: Include report in pull request description
+5. **Re-run After Fixes**: Verify fixes by running analysis again
+6. **Use with Spec**: Always provide feature name for spec cross-reference
+7. **Test All Scenarios**: Execute all critical test scenarios before deployment
+
+---
+
+**Preferred workflow**: `/spec-init` → `/spec-requirements` → `/spec-design` → `/spec-validate` → `/spec-tasks` → `/code` → `/impact-analysis` → `/test` → `/review`