forge-workflow 0.0.1

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (105) hide show
  1. package/.claude/commands/dev.md +314 -0
  2. package/.claude/commands/plan.md +389 -0
  3. package/.claude/commands/premerge.md +179 -0
  4. package/.claude/commands/research.md +42 -0
  5. package/.claude/commands/review.md +442 -0
  6. package/.claude/commands/rollback.md +721 -0
  7. package/.claude/commands/ship.md +134 -0
  8. package/.claude/commands/sonarcloud.md +152 -0
  9. package/.claude/commands/status.md +77 -0
  10. package/.claude/commands/validate.md +237 -0
  11. package/.claude/commands/verify.md +221 -0
  12. package/.claude/rules/greptile-review-process.md +285 -0
  13. package/.claude/rules/workflow.md +105 -0
  14. package/.claude/scripts/greptile-resolve.sh +526 -0
  15. package/.claude/scripts/load-env.sh +32 -0
  16. package/.forge/hooks/check-tdd.js +240 -0
  17. package/.github/PLUGIN_TEMPLATE.json +32 -0
  18. package/.mcp.json.example +12 -0
  19. package/AGENTS.md +169 -0
  20. package/CLAUDE.md +99 -0
  21. package/LICENSE +21 -0
  22. package/README.md +414 -0
  23. package/bin/forge-cmd.js +313 -0
  24. package/bin/forge-validate.js +303 -0
  25. package/bin/forge.js +4228 -0
  26. package/docs/AGENT_INSTALL_PROMPT.md +342 -0
  27. package/docs/ENHANCED_ONBOARDING.md +602 -0
  28. package/docs/EXAMPLES.md +482 -0
  29. package/docs/GREPTILE_SETUP.md +400 -0
  30. package/docs/MANUAL_REVIEW_GUIDE.md +106 -0
  31. package/docs/ROADMAP.md +359 -0
  32. package/docs/SETUP.md +632 -0
  33. package/docs/TOOLCHAIN.md +849 -0
  34. package/docs/VALIDATION.md +363 -0
  35. package/docs/WORKFLOW.md +400 -0
  36. package/docs/planning/PROGRESS.md +396 -0
  37. package/docs/plans/.gitkeep +0 -0
  38. package/docs/plans/2026-02-27-forge-test-suite-v2-decisions.md +21 -0
  39. package/docs/plans/2026-02-27-forge-test-suite-v2-design.md +362 -0
  40. package/docs/plans/2026-02-27-forge-test-suite-v2-tasks.md +343 -0
  41. package/docs/plans/2026-03-02-superpowers-gaps-decisions.md +26 -0
  42. package/docs/plans/2026-03-02-superpowers-gaps-design.md +239 -0
  43. package/docs/plans/2026-03-02-superpowers-gaps-tasks.md +260 -0
  44. package/docs/plans/2026-03-04-agent-command-parity-design.md +163 -0
  45. package/docs/plans/2026-03-04-verify-worktree-cleanup-decisions.md +7 -0
  46. package/docs/plans/2026-03-04-verify-worktree-cleanup-design.md +165 -0
  47. package/docs/plans/2026-03-05-forge-uto-decisions.md +6 -0
  48. package/docs/plans/2026-03-05-forge-uto-design.md +116 -0
  49. package/docs/plans/2026-03-05-forge-uto-tasks.md +244 -0
  50. package/docs/plans/2026-03-10-command-creator-and-eval-decisions.md +52 -0
  51. package/docs/plans/2026-03-10-command-creator-and-eval-design.md +350 -0
  52. package/docs/plans/2026-03-10-command-creator-and-eval-tasks.md +426 -0
  53. package/docs/plans/2026-03-10-stale-workflow-refs-decisions.md +8 -0
  54. package/docs/plans/2026-03-10-stale-workflow-refs-design.md +80 -0
  55. package/docs/plans/2026-03-10-stale-workflow-refs-tasks.md +90 -0
  56. package/docs/plans/2026-03-14-beads-plan-context-decisions.md +9 -0
  57. package/docs/plans/2026-03-14-beads-plan-context-design.md +171 -0
  58. package/docs/plans/2026-03-14-beads-plan-context-tasks.md +160 -0
  59. package/docs/plans/2026-03-14-skill-eval-loop-decisions.md +33 -0
  60. package/docs/plans/2026-03-14-skill-eval-loop-design.md +118 -0
  61. package/docs/plans/2026-03-14-skill-eval-loop-results.md +78 -0
  62. package/docs/plans/2026-03-14-skill-eval-loop-tasks.md +160 -0
  63. package/docs/plans/2026-03-15-agent-command-parity-v2-decisions.md +11 -0
  64. package/docs/plans/2026-03-15-agent-command-parity-v2-design.md +145 -0
  65. package/docs/plans/2026-03-15-agent-command-parity-v2-tasks.md +211 -0
  66. package/docs/research/TEMPLATE.md +292 -0
  67. package/docs/research/advanced-testing.md +297 -0
  68. package/docs/research/agent-permissions.md +167 -0
  69. package/docs/research/dependency-chain.md +328 -0
  70. package/docs/research/forge-workflow-v2.md +550 -0
  71. package/docs/research/plugin-architecture.md +772 -0
  72. package/docs/research/pr4-cli-automation.md +326 -0
  73. package/docs/research/premerge-verify-restructure.md +205 -0
  74. package/docs/research/skills-restructure.md +508 -0
  75. package/docs/research/sonarcloud-perfection-plan.md +166 -0
  76. package/docs/research/sonarcloud-quality-gate.md +184 -0
  77. package/docs/research/superpowers-integration.md +403 -0
  78. package/docs/research/superpowers.md +319 -0
  79. package/docs/research/test-environment.md +519 -0
  80. package/install.sh +1062 -0
  81. package/lefthook.yml +39 -0
  82. package/lib/agents/README.md +198 -0
  83. package/lib/agents/claude.plugin.json +28 -0
  84. package/lib/agents/cline.plugin.json +22 -0
  85. package/lib/agents/codex.plugin.json +19 -0
  86. package/lib/agents/copilot.plugin.json +24 -0
  87. package/lib/agents/cursor.plugin.json +25 -0
  88. package/lib/agents/kilocode.plugin.json +22 -0
  89. package/lib/agents/opencode.plugin.json +20 -0
  90. package/lib/agents/roo.plugin.json +23 -0
  91. package/lib/agents-config.js +2112 -0
  92. package/lib/commands/dev.js +513 -0
  93. package/lib/commands/plan.js +696 -0
  94. package/lib/commands/recommend.js +119 -0
  95. package/lib/commands/ship.js +377 -0
  96. package/lib/commands/status.js +378 -0
  97. package/lib/commands/validate.js +602 -0
  98. package/lib/context-merge.js +359 -0
  99. package/lib/plugin-catalog.js +360 -0
  100. package/lib/plugin-manager.js +166 -0
  101. package/lib/plugin-recommender.js +141 -0
  102. package/lib/project-discovery.js +491 -0
  103. package/lib/setup.js +118 -0
  104. package/lib/workflow-profiles.js +203 -0
  105. package/package.json +115 -0
@@ -0,0 +1,400 @@
1
+ # Greptile Code Review - Branch Protection Setup
2
+
3
+ **✅ Greptile is already working on your repository!**
4
+
5
+ Greptile provides AI-powered code review as a **GitHub App** that automatically analyzes every PR.
6
+
7
+ ---
8
+
9
+ ## Current Status
10
+
11
+ 🎉 **Greptile is Fully Operational!**
12
+
13
+ Your repository has both Greptile features working:
14
+ - ✅ **Greptile Review** (GitHub App) - Provides detailed code review comments
15
+ - ✅ **Greptile Quality Gate** (Workflow) - Enforces minimum score of 4.0/5 before merge
16
+ - ✅ Both integrated into branch protection for master branch
17
+
18
+ ---
19
+
20
+ ## Branch Protection Status
21
+
22
+ ### ✅ Fully Configured!
23
+
24
+ Branch protection for `master` now requires:
25
+
26
+ 1. **Greptile Review** (GitHub App check) - Must pass
27
+ 2. **Greptile Quality Gate (≥4/5)** (Custom workflow) - Must pass with score ≥ 4.0
28
+ 3. **Other Required Checks**: ESLint, CodeQL, dependency-review
29
+ 4. **PR Reviews**: At least 1 approving review required
30
+ 5. **Conversation Resolution**: All review threads must be resolved
31
+
32
+ **Result**: PRs cannot be merged unless:
33
+ - Greptile Review completes successfully
34
+ - Greptile confidence score is at least 4.0/5
35
+ - All other quality checks pass
36
+ - Code has been reviewed and approved
37
+
38
+ ---
39
+
40
+ ## How Greptile Works
41
+
42
+ ### GitHub App Integration
43
+
44
+ - **Automatic**: Runs on every PR (no manual trigger needed)
45
+ - **No Workflow Needed**: Works as a GitHub App, not a GitHub Action
46
+ - **No API Key Required**: Authorized through GitHub App installation
47
+
48
+ ### Review Process
49
+
50
+ ```
51
+ PR created/updated
52
+
53
+ Greptile automatically analyzes code
54
+
55
+ Posts detailed feedback as comments
56
+
57
+ Updates "Greptile Review" check status
58
+
59
+ Pass: ✅ Can merge
60
+ Fail: ❌ Blocked (if required in branch protection)
61
+ ```
62
+
63
+ ### What Greptile Checks
64
+
65
+ - 🐛 **Bugs & Edge Cases**: Potential runtime errors, null pointers, race conditions
66
+ - 🔒 **Security**: Vulnerabilities, injection risks, auth issues
67
+ - 📊 **Code Quality**: Complexity, duplication, naming conventions
68
+ - ⚡ **Performance**: Inefficient algorithms, memory leaks
69
+ - 📝 **Best Practices**: Error handling, type safety, modern patterns
70
+ - 🧪 **Testing**: Missing test coverage, test quality
71
+
72
+ ---
73
+
74
+ ## Understanding Greptile Feedback
75
+
76
+ ### Confidence Score
77
+
78
+ Greptile provides a confidence score (0-5) in the PR description that reflects overall code quality:
79
+
80
+ 📊 **Confidence Score Format**: "Confidence Score: X/5" or "Confidence Score: X out of 5"
81
+ 🎯 **Quality Gate Threshold**: Minimum 4.0/5 required to merge
82
+ ✅ **Detailed inline comments** on specific lines of code
83
+ ✅ **Issue severity** indicators (critical, major, minor)
84
+ ✅ **Actionable suggestions** with example fixes
85
+
86
+ ### Example from Your PR #13
87
+
88
+ Greptile identified and you fixed:
89
+ - ✅ Windows path validation bug
90
+ - ✅ Duplicate function definitions
91
+ - ✅ Incorrect fetch timeout implementation
92
+ - ✅ Security vulnerabilities (command injection)
93
+ - ✅ JSON parse crash issues
94
+ - ✅ Unused variables
95
+
96
+ **Result**: 16/16 issues addressed! 🎉
97
+
98
+ ---
99
+
100
+ ## Addressing Greptile Feedback
101
+
102
+ ### Workflow
103
+
104
+ 1. **Read Comments**
105
+ - Greptile posts inline comments on changed files
106
+ - Each explains the issue and suggests fixes
107
+
108
+ 2. **Fix Issues**
109
+ ```bash
110
+ # Make changes based on feedback
111
+ git add .
112
+ git commit -m "fix: address Greptile feedback"
113
+ git push
114
+ ```
115
+
116
+ 3. **Auto Re-analysis**
117
+ - Greptile automatically reviews again after push
118
+ - Verifies fixes
119
+ - Updates check status
120
+
121
+ 4. **Resolve Conversations**
122
+ - Click "Resolve conversation" on each fixed comment
123
+ - Helps track progress
124
+
125
+ ---
126
+
127
+ ## Branch Protection Behavior
128
+
129
+ ### When "Greptile Review" is Required:
130
+
131
+ ```
132
+ ✅ All issues addressed → Check: SUCCESS → ✅ Can merge
133
+ ❌ Outstanding issues → Check: PENDING → ❌ Blocked
134
+ 🔄 Analysis in progress → Check: PENDING → ❌ Blocked
135
+ ```
136
+
137
+ ### Emergency Override
138
+
139
+ If you **must** merge despite Greptile feedback:
140
+
141
+ 1. **Get approval** from tech lead/architect
142
+ 2. **Document in PR description**:
143
+ ```markdown
144
+ **Emergency Bypass**: Production hotfix for [critical-issue]
145
+ **Greptile Status**: Bypassed
146
+ **Justification**: [detailed reason]
147
+ **Follow-up**: Issue #123 created to address feedback
148
+ ```
149
+ 3. **Temporarily disable branch protection** (admin only)
150
+ 4. **Merge**
151
+ 5. **Re-enable protection immediately**
152
+ 6. **Create follow-up issue** to address Greptile feedback
153
+
154
+ ---
155
+
156
+ ## Configuration
157
+
158
+ ### No Setup Required! ✅
159
+
160
+ Since Greptile is a GitHub App:
161
+
162
+ - ❌ No API keys needed in secrets
163
+ - ❌ No workflow files needed
164
+ - ❌ No manual configuration
165
+
166
+ It just works automatically!
167
+
168
+ ### Managing the GitHub App
169
+
170
+ **View installed apps**:
171
+ ```
172
+ https://github.com/settings/installations
173
+ ```
174
+
175
+ **Repository-specific settings** (admin only):
176
+ ```
177
+ https://github.com/harshanandak/forge/settings/installations
178
+ ```
179
+
180
+ You can:
181
+ - Enable/disable Greptile for specific repos
182
+ - Adjust review frequency
183
+ - Configure notification settings
184
+
185
+ ---
186
+
187
+ ## Customization (Optional)
188
+
189
+ ### Repository Configuration
190
+
191
+ Create `.greptile/config.yml` in repo root:
192
+
193
+ ```yaml
194
+ # Greptile configuration
195
+ review:
196
+ # File patterns to ignore
197
+ exclude:
198
+ - "*.md"
199
+ - "test/**"
200
+ - "docs/**"
201
+ - "*.test.js"
202
+ - "dist/**"
203
+
204
+ # Focus areas (prioritize these checks)
205
+ focus:
206
+ - security
207
+ - bugs
208
+ - performance
209
+
210
+ # Review depth
211
+ depth: thorough # quick, normal, thorough
212
+ ```
213
+
214
+ ### Per-PR Instructions
215
+
216
+ Add comments in PR description to guide Greptile:
217
+
218
+ ```markdown
219
+ @greptile focus on security and performance
220
+ @greptile ignore docs/ and test files
221
+ @greptile be extra strict on src/auth/
222
+ ```
223
+
224
+ ---
225
+
226
+ ## Troubleshooting
227
+
228
+ ### "Greptile Review check not appearing in branch protection"
229
+
230
+ **Cause**: Check hasn't completed at least once on any PR.
231
+
232
+ **Fix**:
233
+ 1. It's currently running on PR #13
234
+ 2. Wait for it to complete
235
+ 3. Then refresh branch protection settings page
236
+ 4. "Greptile Review" should now appear in the list
237
+
238
+ ### "Greptile didn't review my PR"
239
+
240
+ **Possible causes**:
241
+ - GitHub App not installed or disabled
242
+ - PR is a draft (some apps skip drafts)
243
+ - Repository not in allowed list
244
+
245
+ **Fix**:
246
+ 1. Visit: https://github.com/harshanandak/forge/settings/installations
247
+ 2. Verify Greptile is installed and enabled
248
+ 3. Check repository access permissions
249
+ 4. Convert draft to ready for review if applicable
250
+
251
+ ### "How do I request a re-review?"
252
+
253
+ **Methods**:
254
+ 1. **Push new commit** - Triggers automatic re-analysis
255
+ 2. **Comment on PR**: `@greptile please review` or `@greptile recheck`
256
+ 3. **Close and reopen PR** - Forces fresh analysis
257
+
258
+ ### "Can I see why Greptile flagged something?"
259
+
260
+ **Yes!**
261
+ 1. Go to "Files changed" tab in PR
262
+ 2. Find Greptile's comment thread
263
+ 3. Each comment explains:
264
+ - What the issue is
265
+ - Why it's problematic
266
+ - How to fix it
267
+ - Often includes code examples
268
+
269
+ ---
270
+
271
+ ## Best Practices
272
+
273
+ ### 1. Address Feedback Incrementally
274
+
275
+ Don't batch all fixes into one commit:
276
+ - Fix issues as you see them
277
+ - Commit after each logical fix
278
+ - Easier to review and debug
279
+
280
+ ### 2. Use as Learning Tool
281
+
282
+ Greptile explains *why* something is an issue:
283
+ - Read the explanations, don't just apply fixes blindly
284
+ - Share interesting findings with your team
285
+ - Update coding standards based on patterns
286
+
287
+ ### 3. Combine with Human Review
288
+
289
+ | Review Type | What It Catches |
290
+ |-------------|-----------------|
291
+ | 🤖 Greptile | Technical bugs, security, complexity, patterns |
292
+ | 👥 Human | Business logic, UX, architecture, context |
293
+
294
+ **Both are essential!** They catch different types of issues.
295
+
296
+ ### 4. Don't Fight the AI Unnecessarily
297
+
298
+ If Greptile flags something:
299
+ - There's usually a valid reason
300
+ - Read the explanation carefully
301
+ - If you disagree, comment why (helps improve Greptile)
302
+ - Propose alternative if you have a better approach
303
+
304
+ ### 5. Track Common Patterns
305
+
306
+ Notice recurring issues across PRs?
307
+ - Document in coding standards
308
+ - Add to .greptile/config.yml to auto-enforce
309
+ - Share with team in README or CONTRIBUTING.md
310
+ - Consider pre-commit hooks for common issues
311
+
312
+ ---
313
+
314
+ ## Verification Checklist
315
+
316
+ Use this to confirm Greptile is set up correctly:
317
+
318
+ ```
319
+ ✅ Greptile GitHub App is installed
320
+ ✅ Greptile has access to your repository
321
+ ✅ "Greptile Review" check runs on PRs
322
+ ✅ Greptile posts code review comments
323
+ ✅ "Greptile Review" appears in branch protection options
324
+ ✅ "Greptile Review" is selected as required check
325
+ ✅ Branch protection rule is saved
326
+ ✅ Test: Create PR → Greptile reviews → Merge blocked if issues
327
+ ```
328
+
329
+ ---
330
+
331
+ ## FAQ
332
+
333
+ **Q: Does Greptile use a scoring system (like 4.0/5.0)?**
334
+ A: Yes! Greptile Review provides a confidence score (0-5) in the PR description. Our custom Quality Gate workflow enforces a minimum score of 4.0/5 before allowing merges.
335
+
336
+ **Q: Will it review every single commit?**
337
+ A: It reviews at the PR level. Runs when PR is opened and when new commits are pushed.
338
+
339
+ **Q: Does it slow down development?**
340
+ A: No! Reviews typically complete in 1-2 minutes. Runs in parallel with other checks.
341
+
342
+ **Q: Can I disable it for specific PRs?**
343
+ A: Yes, via PR description: `@greptile skip` (but only if not required in branch protection)
344
+
345
+ **Q: Is it free?**
346
+ A: Greptile has free and paid tiers. Check https://greptile.com/pricing for current plans.
347
+
348
+ **Q: Does it replace code review?**
349
+ A: No! It augments human review by catching technical issues, allowing humans to focus on architecture, business logic, and UX.
350
+
351
+ **Q: What languages does it support?**
352
+ A: Most modern languages including JavaScript, TypeScript, Python, Go, Java, Rust, etc.
353
+
354
+ **Q: Can I customize what it checks for?**
355
+ A: Yes, via `.greptile/config.yml` configuration file.
356
+
357
+ ---
358
+
359
+ ## Next Steps
360
+
361
+ 1. ✅ **DONE** - Greptile Review is active and running
362
+ 2. ✅ **DONE** - Greptile Quality Gate (≥4/5) is enforced in branch protection
363
+ 3. ✅ **DONE** - All required checks configured for master branch
364
+ 4. 🎯 **Create new PRs** and watch the quality gate in action
365
+ 5. 📚 **Document** your team's policy for handling Greptile feedback
366
+ 6. 🎉 **Celebrate** improved code quality!
367
+
368
+ ---
369
+
370
+ ## Additional Resources
371
+
372
+ - **Greptile Documentation**: https://docs.greptile.com
373
+ - **GitHub App Settings**: https://github.com/settings/installations
374
+ - **Branch Protection Guide**: [../.github/BRANCH_PROTECTION_GUIDE.md](../.github/BRANCH_PROTECTION_GUIDE.md)
375
+ - **Your PR #13** (example): https://github.com/harshanandak/forge/pull/13
376
+
377
+ ---
378
+
379
+ ## Summary
380
+
381
+ **What Greptile Is:**
382
+ - ✅ GitHub App providing detailed code reviews
383
+ - ✅ Custom Quality Gate workflow enforcing minimum score 4.0/5
384
+ - ✅ AI-powered code analysis on every PR
385
+ - ✅ Detailed, actionable feedback with confidence scores
386
+
387
+ **What's Now Active:**
388
+ - ✅ Greptile Review (GitHub App) is installed and running
389
+ - ✅ Greptile Quality Gate (≥4/5) is enforced in branch protection
390
+ - ✅ PRs to master require score ≥ 4.0/5 to merge
391
+ - ✅ All review comments must be resolved before merge
392
+
393
+ **Result:**
394
+ - 🚀 Higher code quality with enforced standards
395
+ - 🐛 Fewer bugs in production
396
+ - 📊 Objective quality metrics (4.0/5 minimum)
397
+ - 🛡️ Automated security and best practice checks
398
+ - 📚 Team learning from AI feedback
399
+
400
+ Enjoy your new AI code reviewer with quality enforcement! 🤖✨
@@ -0,0 +1,106 @@
1
+ # Manual Review Guide
2
+
3
+ **Purpose**: Structured guidance for manual code review integrated with AI review tools.
4
+
5
+ **Audience**: Developers conducting PR reviews, maintainers configuring review workflows.
6
+
7
+ ---
8
+
9
+ ## Overview
10
+
11
+ Manual review remains essential even with AI-powered tools like Greptile and CodeRabbit. This guide provides a systematic approach to combine human judgment with AI insights for comprehensive code quality.
12
+
13
+ **Review Philosophy**:
14
+ - **AI First, Human Final**: Use AI tools for breadth, manual review for depth
15
+ - **Structured Process**: Follow systematic checklist to avoid missing critical issues
16
+ - **Context Matters**: Apply judgment based on project stage, risk, and complexity
17
+ - **Documentation Required**: Every PR needs clear explanation of changes and reasoning
18
+
19
+ ---
20
+
21
+ ## Review Workflow Integration
22
+
23
+ This guide integrates with the Forge 9-Stage TDD Workflow:
24
+
25
+ ```
26
+ /status → /research → /plan → /dev → /validate → /ship → /review → /merge → /verify
27
+
28
+ You are here
29
+ ```
30
+
31
+ ### When to Use This Guide
32
+
33
+ - **Stage 7 (/review)**: Address ALL PR feedback from GitHub Actions, Greptile, SonarCloud, and manual reviewers
34
+ - **After AI Review**: When Greptile or CodeRabbit has completed initial analysis
35
+ - **Before Merge**: Final verification before approving PR
36
+ - **Post-Merge**: Documentation verification in /verify stage
37
+
38
+
39
+
40
+ ## Part 1: AI Review Tools Best Practices
41
+
42
+ ### Greptile - Semantic Understanding
43
+ - Use .claude/scripts/greptile-resolve.sh for systematic thread handling
44
+ - Always reply and resolve threads after fixes
45
+ - See .claude/rules/greptile-review-process.md for detailed workflow
46
+
47
+ ### CodeRabbit - Multi-Model Review
48
+ - Address security issues immediately
49
+ - Consider performance suggestions with benchmarks
50
+ - Apply style suggestions for consistency
51
+
52
+ ### SonarCloud - Static Analysis
53
+ - Coverage ≥80% on new code
54
+ - 0 security hotspots unreviewed
55
+ - Use /sonarcloud skill for PR-specific issues
56
+
57
+ ---
58
+
59
+ ## Part 2: Manual Review Checklist
60
+
61
+ ### 1. Functional Correctness
62
+ - ☐ Code matches PR description
63
+ - ☐ Edge cases handled
64
+ - ☐ Error messages clear
65
+ - ☐ Invalid input handled gracefully
66
+
67
+ ### 2. Security (OWASP Top 10)
68
+ - ☐ Authorization before sensitive ops
69
+ - ☐ Data encrypted at rest/transit
70
+ - ☐ SQL queries parameterized
71
+ - ☐ No code injection risks
72
+
73
+ ### 3. Testing Quality
74
+ - ☐ Tests for new code
75
+ - ☐ Edge cases covered
76
+ - ☐ TDD compliance (test commits before feat commits)
77
+
78
+ ### 4. Code Quality
79
+ - ☐ Self-documenting code
80
+ - ☐ Single responsibility functions
81
+ - ☐ No duplication (DRY)
82
+ - ☐ Clear organization
83
+
84
+ ### 5. Performance
85
+ - ☐ Efficient algorithms
86
+ - ☐ Optimized queries
87
+ - ☐ No memory leaks
88
+
89
+ ### 6. Documentation
90
+ - ☐ Public APIs documented
91
+ - ☐ PR explains why, not just what
92
+ - ☐ README updated if needed
93
+
94
+ ---
95
+
96
+ ## Summary
97
+
98
+ **Manual Review Essentials**:
99
+ - Use AI tools for pattern detection
100
+ - Apply human judgment for context
101
+ - Follow systematic checklist
102
+ - Verify security, testing, documentation
103
+
104
+ **Integration**: Stage 7 (/review) → Address ALL feedback systematically
105
+
106
+ **Key**: Trust AI for breadth, humans for depth. Best results combine both.