[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M75GDMS4CX35CQM63H020G2K":3,"skill-same:01M75GDMS4CX35CQM63H020G2K":54},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":31,"commitSha":32,"skillDirectory":33,"forksCount":34,"repositoryHtmlUrl":35,"repositoryHomepageUrl":6,"ownerHtmlUrl":36,"ownerBlogUrl":37,"ownerDescription":38,"updatedAt":39,"install":40,"resources":53},"01M75GDMS4CX35CQM63H020G2K","solo-review",null,"Final code review and quality gate — run tests, check coverage, audit security, verify acceptance criteria from spec, and generate ship-ready report. Use when user says \"review code\", \"quality check\", \"is it ready to ship\", \"final review\", or after \u002Fdeploy completes. Do NOT use for planning (use \u002Fplan) or building (use \u002Fbuild).","javimosch\u002Fopen-claw-skills","javimosch","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F12900508?v=4","01M6D3F2AS6PV75ET9PZ1QGP8J","Javier Leandro Arancibia","01M75B8NCTAJG9WH5YNZH7C3D8",0,"MIT","https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fopen-claw-skills\u002Fblob\u002Fmain\u002Fskills\u002Ffortunto2\u002Fsolo-review\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nname: solo-review\ndescription: Final code review and quality gate — run tests, check coverage, audit security, verify acceptance criteria from spec, and generate ship-ready report. Use when user says \"review code\", \"quality check\", \"is it ready to ship\", \"final review\", or after \u002Fdeploy completes. Do NOT use for planning (use \u002Fplan) or building (use \u002Fbuild).\nlicense: MIT\nmetadata:\n  author: fortunto2\n  version: \"1.1.1\"\n  openclaw:\n    emoji: \"🔎\"\nallowed-tools: Read, Grep, Bash, Glob, Write, Edit, mcp__solograph__session_search, mcp__solograph__project_code_search, mcp__solograph__codegraph_query, mcp__solograph__codegraph_explain, mcp__solograph__web_search, mcp__context7__resolve-library-id, mcp__context7__query-docs\nargument-hint: \"[focus-area]\"\n---\n\n# \u002Freview\n\nThis skill is self-contained — follow the instructions below instead of delegating to external review skills (superpowers, etc.) or spawning Task subagents. Run all checks directly.\n\nFinal quality gate before shipping. Runs tests, checks security, verifies acceptance criteria from spec.md, audits code quality, and generates a ship-ready report with go\u002Fno-go verdict.\n\n## When to use\n\nAfter `\u002Fdeploy` (or `\u002Fbuild` if deploying manually). This is the quality gate.\n\nPipeline: `\u002Fdeploy` → **`\u002Freview`**\n\nCan also be used standalone: `\u002Freview` on any project to audit code quality.\n\n## MCP Tools (use if available)\n\n- `session_search(query)` — find past review patterns and common issues\n- `project_code_search(query, project)` — find similar code patterns across projects\n- `codegraph_query(query)` — check dependencies, imports, unused code\n\nIf MCP tools are not available, fall back to Glob + Grep + Read.\n\n## Pre-flight Checks\n\n### 1. Architecture overview (if MCP available)\n```\ncodegraph_explain(project=\"{project name}\")\n```\nReturns: stack, languages, directory layers, key patterns, top dependencies, hub files. Use this to detect stack and understand project structure.\n\n### 2. Essential docs (parallel reads)\n- `CLAUDE.md` — architecture, Do\u002FDon't rules\n- `docs\u002Fplan\u002F*\u002Fspec.md` — acceptance criteria to verify (REQUIRED)\n- `docs\u002Fplan\u002F*\u002Fplan.md` — task completion status (REQUIRED)\n- `docs\u002Fworkflow.md` — TDD policy, quality standards, **integration testing commands** (if exists)\n\n**Do NOT read source code at this stage.** Only docs.\n\n### 3. Detect stack\nUse stack from `codegraph_explain` response (or `CLAUDE.md` if no MCP) to choose tools:\n- Next.js → `npm run build`, `npm test`, `npx next lint`\n- Python → `uv run pytest`, `uv run ruff check`\n- Swift → `swift test`, `swiftlint`\n- Kotlin → `.\u002Fgradlew test`, `.\u002Fgradlew lint`\n\n### 4. Smart source code loading (for code quality spot check)\n\n**Do NOT read random source files.** Use the graph to find the most important code:\n\n```\ncodegraph_query(\"MATCH (f:File {project: '{name}'})-[e]-() RETURN f.path, COUNT(e) AS edges ORDER BY edges DESC LIMIT 5\")\n```\n\nRead only the top 3-5 hub files (most connected = most impactful). For security checks, use Grep with narrow patterns (`sk_live`, `password\\s*=`) — not full file reads.\n\n## Review Dimensions\n\n**Makefile convention:** If `Makefile` exists in project root, **always prefer `make` targets** over raw commands. Use `make test` instead of `npm test`, `make lint` instead of `pnpm lint`, `make build` instead of `pnpm build`. Run `make help` (or read Makefile) to discover available targets including integration tests.\n\nRun all 12 dimensions in sequence. Report findings per dimension.\n\n### 1. Test Suite\n\nRun the full test suite (prefer `make test` if Makefile exists):\n```bash\n# If Makefile exists — use it\nmake test 2>&1 || true\n\n# Fallback: Next.js \u002F Node\nnpm test -- --coverage 2>&1 || true\n\n# Python\nuv run pytest --tb=short -q 2>&1 || true\n\n# Swift\nswift test 2>&1 || true\n```\n\nReport:\n- Total tests: pass \u002F fail \u002F skip\n- Coverage percentage (if available)\n- Any failing tests with file:line references\n\n**Integration tests** — if `docs\u002Fworkflow.md` has an \"Integration Testing\" section, run the specified commands:\n- Execute the CLI\u002Fintegration commands listed there\n- Verify exit code 0 and expected output format\n- Report: command run, exit code, pass\u002Ffail\n\n### 2. Linter & Type Check\n\n```bash\n# Next.js\npnpm lint 2>&1 || true\npnpm tsc --noEmit 2>&1 || true\n\n# Python\nuv run ruff check . 2>&1 || true\nuv run ty check . 2>&1 || true\n\n# Swift\nswiftlint lint --strict 2>&1 || true\n\n# Kotlin\n.\u002Fgradlew detekt 2>&1 || true\n.\u002Fgradlew ktlintCheck 2>&1 || true\n```\n\nReport: warnings count, errors count, top issues.\n\n### 3. Build Verification\n\n```bash\n# Next.js\nnpm run build 2>&1 || true\n\n# Python\nuv run python -m py_compile src\u002F**\u002F*.py 2>&1 || true\n\n# Astro\nnpm run build 2>&1 || true\n```\n\nReport: build success\u002Ffailure, any warnings.\n\n### 4. Security Audit\n\n**Dependency vulnerabilities:**\n```bash\n# Node\nnpm audit --audit-level=moderate 2>&1 || true\n\n# Python\nuv run pip-audit 2>&1 || true\n```\n\n**Code-level checks** (Grep for common issues):\n- Hardcoded secrets: `grep -rn \"sk_live\\|sk_test\\|password\\s*=\\s*['\\\"]\" src\u002F app\u002F lib\u002F`\n- SQL injection: look for string concatenation in queries\n- XSS: look for `dangerouslySetInnerHTML` without sanitization\n- Exposed env vars: check `.gitignore` includes `.env*`\n\nReport: vulnerabilities found, severity levels.\n\n### 5. Acceptance Criteria Verification\n\nRead `docs\u002Fplan\u002F*\u002Fspec.md` and check each acceptance criterion:\n\nFor each `- [ ]` criterion in spec.md:\n1. Search codebase for evidence it was implemented.\n2. Check if related tests exist.\n3. Mark as verified or flag as missing.\n\n**Update spec.md checkboxes.** After verifying each criterion, use Edit tool to change `- [ ]` to `- [x]` in spec.md. Leaving verified criteria unchecked causes staleness across pipeline runs — check them off as you go.\n\n```\nAcceptance Criteria:\n  - [x] User can sign up with email — found in app\u002Fauth\u002Fsignup\u002Fpage.tsx + test\n  - [x] Dashboard shows project list — found in app\u002Fdashboard\u002Fpage.tsx\n  - [ ] Stripe checkout works — route exists but no test coverage\n```\n\nAfter updating checkboxes, commit: `git add docs\u002Fplan\u002F*\u002Fspec.md && git commit -m \"docs: update spec checkboxes (verified by review)\"`\n\n### 6. Code Quality Spot Check\n\nRead 3-5 key files (entry points, API routes, main components):\n- Check for TODO\u002FFIXME\u002FHACK comments that should be resolved\n- Check for console.log\u002Fprint statements left in production code\n- Check for proper error handling (try\u002Fcatch, error boundaries)\n- Check for proper loading\u002Ferror states in UI components\n\nReport specific file:line references for any issues found.\n\n### 7. Plan Completion Check\n\nRead `docs\u002Fplan\u002F*\u002Fplan.md`:\n- Count completed tasks `[x]` vs total tasks\n- Flag any `[ ]` or `[~]` tasks still remaining\n- Verify all phase checkpoints have SHAs\n\n### 8. Production Logs (if deployed)\n\nIf the project has been deployed (deploy URL in CLAUDE.md, or `.solo\u002Fstates\u002Fdeploy` exists if pipeline state directory is present), **check production logs for runtime errors**.\n\nRead the `logs` field from the stack YAML (`templates\u002Fstacks\u002F{stack}.yaml`) to get platform-specific commands.\n\n**Vercel (Next.js):**\n```bash\nvercel logs --output=short 2>&1 | tail -50\n```\nLook for: `Error`, `FUNCTION_INVOCATION_FAILED`, `504`, unhandled rejections, hydration mismatches.\n\n**Cloudflare Workers:**\n```bash\nwrangler tail --format=pretty 2>&1 | head -50\n```\nLook for: uncaught exceptions, D1 errors, R2 access failures.\n\n**Fly.io (Python API):**\n```bash\nfly logs --app {name} 2>&1 | tail -50\n```\nLook for: `ERROR`, `CRITICAL`, OOM, connection refused, unhealthy instances.\n\n**Supabase Edge Functions:**\n```bash\nsupabase functions logs --scroll 2>&1 | tail -30\n```\n\n**iOS (TestFlight):**\n- Check App Store Connect → TestFlight → Crashes\n- If local device: `log stream --predicate 'subsystem == \"com.{org}.{name}\"'`\n\n**Android:**\n```bash\nadb logcat '*:E' --format=time 2>&1 | tail -30\n```\n- Check Google Play Console → Android vitals → Crashes & ANRs\n\n**If no deploy yet:** skip this dimension, note in report as \"N\u002FA — not deployed\".\n\n**If logs show errors:**\n- Classify: startup crash vs runtime error vs intermittent\n- Add as FIX FIRST issues in the report\n- Include exact log lines as evidence\n\nReport:\n- Log source checked (platform, command used)\n- Errors found: count + severity\n- Error patterns (recurring vs one-off)\n- Status: CLEAN \u002F WARN \u002F ERRORS\n\n### 9. Dev Principles Compliance\n\nCheck adherence to dev principles. Look for `templates\u002Fprinciples\u002Fdev-principles.md` (bundled with this skill), or check CLAUDE.md or project docs for architecture and coding conventions.\n\nRead the dev principles file, then spot-check 3-5 key source files for violations:\n\n**SOLID:**\n- **SRP** — any god-class\u002Fgod-module doing auth + profile + email + notifications? Flag bloated files (>300 LOC with mixed responsibilities).\n- **DIP** — are services injected or hardcoded? Look for `new ConcreteService()` inside business logic instead of dependency injection.\n\n**DRY vs Rule of Three:**\n- Search for duplicated logic blocks (Grep for identical function signatures across files).\n- But don't flag 2-3 similar lines — duplication is OK until a pattern emerges.\n\n**KISS:**\n- Over-engineered abstractions for one-time operations?\n- Feature flags or backward-compat shims where a simple change would do?\n- Helpers\u002Futilities used only once?\n\n**Schemas-First (SGR):**\n- Are Pydantic\u002FZod schemas defined before logic? Or is raw data passed around?\n- Are API responses typed (not `any` \u002F `dict`)?\n- Validation at boundaries (user input, external APIs)?\n\n**Clean Architecture:**\n- Do dependencies point inward? Business logic should not import from UI\u002Fframework layer.\n- Is business logic framework-independent?\n\n**Error Handling:**\n- Fail-fast on invalid inputs? Or silent swallowing of errors?\n- User-facing errors are friendly? Internal errors have stack traces?\n\nReport:\n- Principles followed: list key ones observed\n- Violations found: with file:line references\n- Severity: MINOR (style) \u002F MAJOR (architecture) \u002F CRITICAL (data loss risk)\n\n### 10. Commit Quality\n\nCheck git history for the current track\u002Ffeature:\n\n```bash\ngit log --oneline --since=\"1 week ago\" 2>&1 | head -30\n```\n\n**Conventional commits format:**\n- Each commit follows `\u003Ctype>(\u003Cscope>): \u003Cdescription>` pattern\n- Types: `feat`, `fix`, `refactor`, `test`, `docs`, `chore`, `perf`, `style`\n- Flag: generic messages (\"fix\", \"update\", \"wip\", \"changes\"), missing type prefix, too-long titles (>72 chars)\n\n**Atomicity:**\n- Each commit = one logical change? Or monster commits with 20 files across unrelated features?\n- Revert-friendly? Could you `git revert` a single commit without side effects?\n\n**SHAs in plan.md:**\n- Check that completed tasks have `\u003C!-- sha:abc1234 -->` comments\n- Check that phase checkpoints have `\u003C!-- checkpoint:abc1234 -->`\n\n```bash\ngrep -c \"sha:\" docs\u002Fplan\u002F*\u002Fplan.md 2>\u002Fdev\u002Fnull || echo \"No SHAs found\"\n```\n\n**Pre-commit hooks:**\n\nRead the stack YAML `pre_commit` field to know what system is expected (husky\u002Fpre-commit\u002Flefthook) and what it should run (linter + formatter + type-checker). Then verify:\n\n```bash\n# Detect what's configured\n[ -f .husky\u002Fpre-commit ] && echo \"husky\" || [ -f .pre-commit-config.yaml ] && echo \"pre-commit\" || [ -f lefthook.yml ] && echo \"lefthook\" || echo \"none\"\n```\n\n- **Hooks installed?** Check config files exist AND hooks are wired (`core.hooksPath` for husky, `.git\u002Fhooks\u002Fpre-commit` for pre-commit\u002Flefthook).\n- **Hooks match stack?** Compare detected system with stack YAML `pre_commit` field. Flag mismatch.\n- **`--no-verify` bypasses?** Check if recent commits show signs of skipped hooks (e.g., lint violations that should've been caught). Flag as WARN.\n- **Not configured?** Flag as WARN recommendation — stack YAML expects `{pre_commit}` but nothing found.\n\nReport:\n- Total commits: {N}\n- Conventional format: {N}\u002F{M} compliant\n- Atomic commits: YES \u002F NO (with examples of violations)\n- Plan SHAs: {N}\u002F{M} tasks have SHAs\n- Pre-commit hooks: {ACTIVE \u002F NOT INSTALLED \u002F NOT CONFIGURED} (expected: {stack pre_commit})\n\n### 11. Documentation Freshness\n\nCheck that project documentation is up-to-date with the code.\n\n**Required files check:**\n```bash\nls -la CLAUDE.md README.md docs\u002Fprd.md docs\u002Fworkflow.md 2>&1\n```\n\n**CLAUDE.md:**\n- Does it reflect current tech stack, commands, directory structure?\n- Are recently added features\u002Fendpoints documented?\n- Grep for outdated references (old package names, removed files):\n  ```bash\n  # Check that files mentioned in CLAUDE.md actually exist\n  grep -oP '`[a-zA-Z0-9_.\u002F-]+\\.(ts|py|swift|kt|md)`' CLAUDE.md | while read f; do [ ! -f \"$f\" ] && echo \"MISSING: $f\"; done\n  ```\n\n**README.md:**\n- Does it have setup\u002Frun\u002Ftest\u002Fdeploy instructions?\n- Are the commands actually runnable?\n\n**docs\u002Fprd.md:**\n- Do features match what was actually built?\n- Are metrics and success criteria defined?\n\n**AICODE- comments:**\n```bash\ngrep -rn \"AICODE-TODO\" src\u002F app\u002F lib\u002F 2>\u002Fdev\u002Fnull | head -10\ngrep -rn \"AICODE-ASK\" src\u002F app\u002F lib\u002F 2>\u002Fdev\u002Fnull | head -10\n```\n- Flag unresolved `AICODE-TODO` items that were completed but not cleaned up\n- Flag unanswered `AICODE-ASK` questions\n- Check for `AICODE-NOTE` on complex\u002Fnon-obvious logic\n\n**Dead code check:**\n- Unused imports (linter should catch, but verify)\n- Orphaned files not imported anywhere\n- If `knip` available (Next.js): `pnpm knip 2>&1 | head -30`\n\nReport:\n- CLAUDE.md: CURRENT \u002F STALE \u002F MISSING\n- README.md: CURRENT \u002F STALE \u002F MISSING\n- docs\u002Fprd.md: CURRENT \u002F STALE \u002F MISSING\n- docs\u002Fworkflow.md: CURRENT \u002F STALE \u002F MISSING\n- AICODE-TODO unresolved: {N}\n- AICODE-ASK unanswered: {N}\n- Dead code: {files\u002Fexports found}\n\n### 12. Visual\u002FE2E Testing\n\nIf browser tools or device tools are available, run a visual smoke test.\n\n**Web projects (Playwright MCP or browser tools):**\n1. Start dev server (use `dev_server.command` from stack YAML, e.g. `pnpm dev`)\n2. Use Playwright MCP tools (or browser-use skill) to navigate to the main page\n3. Verify it loads without console errors, hydration mismatches, or React errors\n4. Navigate to 2-3 key pages (based on spec.md features)\n5. Take screenshots at desktop (1280px) and mobile (375px) viewports\n6. Look for broken images, missing styles, layout overflow\n\n**iOS projects (simulator):**\n1. Build for simulator: `xcodebuild -scheme {Name} -sdk iphonesimulator build`\n2. Install and launch on booted simulator\n3. Take screenshot of main screen\n4. Check simulator logs for crashes or assertion failures\n\n**Android projects (emulator):**\n1. Build debug APK: `.\u002Fgradlew assembleDebug`\n2. Install and launch on emulator\n3. Take screenshot of main activity\n4. Check logcat for crashes or ANRs: `adb logcat '*:E' --format=time -d 2>&1 | tail -20`\n\n**If tools are not available:** skip this dimension, note as \"N\u002FA — no browser\u002Fdevice tools\" in the report. Visual testing is never a blocker for SHIP verdict on its own.\n\nReport:\n- Platform tested: {browser \u002F simulator \u002F emulator \u002F N\u002FA}\n- Pages\u002Fscreens checked: {N}\n- Console errors: {N}\n- Visual issues: {NONE \u002F list}\n- Responsive: {PASS \u002F issues found}\n- Status: {PASS \u002F WARN \u002F FAIL \u002F N\u002FA}\n\n## Review Report\n\nGenerate the final report:\n\n```\nCode Review: {project-name}\nDate: {YYYY-MM-DD}\n\n## Verdict: {SHIP \u002F FIX FIRST \u002F BLOCK}\n\n### Summary\n{1-2 sentence overall assessment}\n\n### Tests\n- Total: {N} | Pass: {N} | Fail: {N} | Skip: {N}\n- Coverage: {N}%\n- Status: {PASS \u002F FAIL}\n\n### Linter\n- Errors: {N} | Warnings: {N}\n- Status: {PASS \u002F WARN \u002F FAIL}\n\n### Build\n- Status: {PASS \u002F FAIL}\n- Warnings: {N}\n\n### Security\n- Vulnerabilities: {N} (critical: {N}, high: {N}, moderate: {N})\n- Hardcoded secrets: {NONE \u002F FOUND}\n- Status: {PASS \u002F WARN \u002F FAIL}\n\n### Acceptance Criteria\n- Verified: {N}\u002F{M}\n- Missing: {list}\n- Status: {PASS \u002F PARTIAL \u002F FAIL}\n\n### Plan Progress\n- Tasks: {N}\u002F{M} complete\n- Phases: {N}\u002F{M} complete\n- Status: {COMPLETE \u002F IN PROGRESS}\n\n### Production Logs\n- Platform: {Vercel \u002F Cloudflare \u002F Fly.io \u002F N\u002FA}\n- Errors: {N} | Warnings: {N}\n- Status: {CLEAN \u002F WARN \u002F ERRORS \u002F N\u002FA}\n\n### Dev Principles\n- SOLID: {PASS \u002F violations found}\n- Schemas-first: {YES \u002F raw data found}\n- Error handling: {PASS \u002F issues found}\n- Status: {PASS \u002F WARN \u002F FAIL}\n\n### Commits\n- Total: {N} | Conventional: {N}\u002F{M}\n- Atomic: {YES \u002F NO}\n- Plan SHAs: {N}\u002F{M}\n- Status: {PASS \u002F WARN \u002F FAIL}\n\n### Documentation\n- CLAUDE.md: {CURRENT \u002F STALE \u002F MISSING}\n- README.md: {CURRENT \u002F STALE \u002F MISSING}\n- AICODE-TODO unresolved: {N}\n- Dead code: {NONE \u002F found}\n- Status: {PASS \u002F WARN \u002F FAIL}\n\n### Visual Testing\n- Platform: {browser \u002F simulator \u002F emulator \u002F N\u002FA}\n- Pages\u002Fscreens: {N}\n- Console errors: {N}\n- Visual issues: {NONE \u002F list}\n- Status: {PASS \u002F WARN \u002F FAIL \u002F N\u002FA}\n\n### Issues Found\n1. [{severity}] {description} — {file:line}\n2. [{severity}] {description} — {file:line}\n\n### Recommendations\n- {actionable recommendation}\n- {actionable recommendation}\n```\n\n**Verdict logic:**\n- **SHIP**: All tests pass, no security issues, acceptance criteria met, build succeeds, production logs clean, docs current, commits atomic, no critical visual issues\n- **FIX FIRST**: Minor issues (warnings, partial criteria, low-severity vulns, intermittent log errors, stale docs, non-conventional commits, minor SOLID violations, minor visual issues like layout overflow) — list what to fix\n- **BLOCK**: Failing tests, security vulnerabilities, missing critical features, production crashes in logs, missing CLAUDE.md\u002FREADME.md, critical architecture violations, app crashes on launch (simulator\u002Femulator) — do not ship\n\n## Post-Verdict: CLAUDE.md Revision\n\nAfter the verdict report, revise the project's CLAUDE.md to keep it lean and useful for future agents.\n\n### Steps:\n\n1. **Read CLAUDE.md** and check size: `wc -c CLAUDE.md`\n2. **Add learnings from this review:**\n   - New Do\u002FDon't rules discovered during review\n   - Updated commands, workflows, or architecture decisions\n   - Fixed issues or gotchas worth remembering\n   - Stack\u002Fdependency changes (new packages, removed deps)\n3. **If over 40,000 characters — trim ruthlessly:**\n   - Collapse completed phase\u002Fmilestone histories into one line each\n   - Remove verbose explanations — keep terse, actionable notes\n   - Remove duplicate info (same thing explained in multiple sections)\n   - Remove historical migration notes, old debugging context\n   - Remove examples that are obvious from code or covered by skill\u002Fdoc files\n   - Remove outdated troubleshooting for resolved issues\n4. **Verify result ≤ 40,000 characters** — if still over, cut least actionable content\n5. **Write updated CLAUDE.md**, update \"Last updated\" date\n\n### Priority (keep → cut):\n1. **ALWAYS KEEP:** Tech stack, directory structure, Do\u002FDon't rules, common commands, architecture decisions\n2. **KEEP:** Workflow instructions, troubleshooting for active issues, key file references\n3. **CONDENSE:** Phase histories (one line each), detailed examples, tool\u002FMCP listings\n4. **CUT FIRST:** Historical notes, verbose explanations, duplicated content, resolved issues\n\n### Rules:\n- Never remove Do\u002FDon't sections — critical guardrails\n- Preserve overall section structure and ordering\n- Every line must earn its place: \"would a future agent need this to do their job?\"\n- Commit the update: `git add CLAUDE.md && git commit -m \"docs: revise CLAUDE.md (post-review)\"`\n\n## AFTER CLAUDE.md revision — output signal EXACTLY ONCE:\n\nOutput pipeline signal ONLY if pipeline state directory (`.solo\u002Fstates\u002F`) exists.\n\n**Output the signal tag ONCE and ONLY ONCE.** Do not repeat it. The pipeline detects the first occurrence.\n\n**If SHIP:** output this exact line (once):\n```\n\u003Csolo:done\u002F>\n```\n\n**If FIX FIRST or BLOCK:**\n1. Open plan.md and APPEND a new phase with fix tasks (one `- [ ] Task` per issue found)\n2. Change plan.md status from `[x] Complete` to `[~] In Progress`\n3. Commit: `git add docs\u002Fplan\u002F && git commit -m \"fix: add review fix tasks\"`\n4. Output this exact line (once):\n```\n\u003Csolo:redo\u002F>\n```\n\nThe pipeline reads these tags and handles all marker files automatically. You do NOT need to create or delete any marker files yourself.\n**Output the signal tag once — the pipeline detects the first occurrence.**\n\n## Error Handling\n\n### Tests won't run\n**Cause:** Missing dependencies or test config.\n**Fix:** Run `npm install` \u002F `uv sync`, check test config exists (jest.config, pytest.ini).\n\n### Linter not configured\n**Cause:** No linter config file found.\n**Fix:** Note as a recommendation in the report, not a blocker.\n\n### Build fails\n**Cause:** Type errors, import issues, missing env vars.\n**Fix:** Report specific errors. This is a BLOCK verdict — must fix before shipping.\n\n## Two-Stage Review Pattern\n\nWhen reviewing significant work, use two stages:\n\n**Stage 1 — Spec Compliance:**\n- Does the implementation match spec.md requirements?\n- Are all acceptance criteria actually met (not just claimed)?\n- Any deviations from the plan? If so, are they justified improvements or problems?\n\n**Stage 2 — Code Quality:**\n- Architecture patterns, error handling, type safety\n- Test coverage and test quality\n- Security and performance\n- Code organization and maintainability\n\n## Verification Gate\n\n**No verdict without fresh evidence.**\n\nBefore writing any verdict (SHIP\u002FFIX\u002FBLOCK):\n1. **Run** the actual test\u002Fbuild\u002Flint commands (not cached results).\n2. **Read** full output — exit codes, pass\u002Ffail counts, error messages.\n3. **Confirm** the output matches your claim.\n4. **Only then** write the verdict with evidence.\n\nNever write \"tests should pass\" — run them and show the output.\n\n## Rationalizations Catalog\n\n| Thought | Reality |\n|---------|---------|\n| \"Tests were passing earlier\" | Run them NOW. Code changed since then. |\n| \"It's just a warning\" | Warnings become bugs. Report them. |\n| \"The build worked locally\" | Check the platform too. Environment differences matter. |\n| \"Security scan is overkill\" | One missed secret = data breach. Always scan. |\n| \"Good enough to ship\" | Quantify \"good enough\". Show the numbers. |\n| \"I already checked this\" | Fresh evidence only. Stale checks are worthless. |\n\n## Critical Rules\n\n1. **Run all checks** — do not skip dimensions even if project seems simple.\n2. **Be specific** — always include file:line references for issues.\n3. **Verdict must be justified** — every SHIP\u002FFIX\u002FBLOCK needs evidence from actual commands.\n4. **Don't auto-fix code** — report issues and add fix tasks to plan.md. Let `\u002Fbuild` fix them. Review only modifies plan.md, never source code.\n5. **Check acceptance criteria** — spec.md is the source of truth for \"done\".\n6. **Security is non-negotiable** — any hardcoded secret = BLOCK.\n7. **Fresh evidence only** — run commands before making claims. Never rely on memory.\n","# \u002Freview\n\nThis skill is self-contained — follow the instructions below instead of delegating to external review skills (superpowers, etc.) or spawning Task subagents. Run all checks directly.\n\nFinal quality gate before shipping. Runs tests, checks security, verifies acceptance criteria from spec.md, audits code quality, and generates a ship-ready report with go\u002Fno-go verdict.\n\n## When to use\n\nAfter `\u002Fdeploy` (or `\u002Fbuild` if deploying manually). This is the quality gate.\n\nPipeline: `\u002Fdeploy` → **`\u002Freview`**\n\nCan also be used standalone: `\u002Freview` on any project to audit code quality.\n\n## MCP Tools (use if available)\n\n- `session_search(query)` — find past review patterns and common issues\n- `project_code_search(query, project)` — find similar code patterns across projects\n- `codegraph_query(query)` — check dependencies, imports, unused code\n\nIf MCP tools are not available, fall back to Glob + Grep + Read.\n\n## Pre-flight Checks\n\n### 1. Architecture overview (if MCP available)\n```\ncodegraph_explain(project=\"{project name}\")\n```\nReturns: stack, languages, directory layers, key patterns, top dependencies, hub files. Use this to detect stack and understand project structure.\n\n### 2. Essential docs (parallel reads)\n- `CLAUDE.md` — architecture, Do\u002FDon't rules\n- `docs\u002Fplan\u002F*\u002Fspec.md` — acceptance criteria to verify (REQUIRED)\n- `docs\u002Fplan\u002F*\u002Fplan.md` — task completion status (REQUIRED)\n- `docs\u002Fworkflow.md` — TDD policy, quality standards, **integration testing commands** (if exists)\n\n**Do NOT read source code at this stage.** Only docs.\n\n### 3. Detect stack\nUse stack from `codegraph_explain` response (or `CLAUDE.md` if no MCP) to choose tools:\n- Next.js → `npm run build`, `npm test`, `npx next lint`\n- Python → `uv run pytest`, `uv run ruff check`\n- Swift → `swift test`, `swiftlint`\n- Kotlin → `.\u002Fgradlew test`, `.\u002Fgradlew lint`\n\n### 4. Smart source code loading (for code quality spot check)\n\n**Do NOT read random source files.** Use the graph to find the most important code:\n\n```\ncodegraph_query(\"MATCH (f:File {project: '{name}'})-[e]-() RETURN f.path, COUNT(e) AS edges ORDER BY edges DESC LIMIT 5\")\n```\n\nRead only the top 3-5 hub files (most connected = most impactful). For security checks, use Grep with narrow patterns (`sk_live`, `password\\s*=`) — not full file reads.\n\n## Review Dimensions\n\n**Makefile convention:** If `Makefile` exists in project root, **always prefer `make` targets** over raw commands. Use `make test` instead of `npm test`, `make lint` instead of `pnpm lint`, `make build` instead of `pnpm build`. Run `make help` (or read Makefile) to discover available targets including integration tests.\n\nRun all 12 dimensions in sequence. Report findings per dimension.\n\n### 1. Test Suite\n\nRun the full test suite (prefer `make test` if Makefile exists):\n```bash\n# If Makefile exists — use it\nmake test 2>&1 || true\n\n# Fallback: Next.js \u002F Node\nnpm test -- --coverage 2>&1 || true\n\n# Python\nuv run pytest --tb=short -q 2>&1 || true\n\n# Swift\nswift test 2>&1 || true\n```\n\nReport:\n- Total tests: pass \u002F fail \u002F skip\n- Coverage percentage (if available)\n- Any failing tests with file:line references\n\n**Integration tests** — if `docs\u002Fworkflow.md` has an \"Integration Testing\" section, run the specified commands:\n- Execute the CLI\u002Fintegration commands listed there\n- Verify exit code 0 and expected output format\n- Report: command run, exit code, pass\u002Ffail\n\n### 2. Linter & Type Check\n\n```bash\n# Next.js\npnpm lint 2>&1 || true\npnpm tsc --noEmit 2>&1 || true\n\n# Python\nuv run ruff check . 2>&1 || true\nuv run ty check . 2>&1 || true\n\n# Swift\nswiftlint lint --strict 2>&1 || true\n\n# Kotlin\n.\u002Fgradlew detekt 2>&1 || true\n.\u002Fgradlew ktlintCheck 2>&1 || true\n```\n\nReport: warnings count, errors count, top issues.\n\n### 3. Build Verification\n\n```bash\n# Next.js\nnpm run build 2>&1 || true\n\n# Python\nuv run python -m py_compile src\u002F**\u002F*.py 2>&1 || true\n\n# Astro\nnpm run build 2>&1 || true\n```\n\nReport: build success\u002Ffailure, any warnings.\n\n### 4. Security Audit\n\n**Dependency vulnerabilities:**\n```bash\n# Node\nnpm audit --audit-level=moderate 2>&1 || true\n\n# Python\nuv run pip-audit 2>&1 || true\n```\n\n**Code-level checks** (Grep for common issues):\n- Hardcoded secrets: `grep -rn \"sk_live\\|sk_test\\|password\\s*=\\s*['\\\"]\" src\u002F app\u002F lib\u002F`\n- SQL injection: look for string concatenation in queries\n- XSS: look for `dangerouslySetInnerHTML` without sanitization\n- Exposed env vars: check `.gitignore` includes `.env*`\n\nReport: vulnerabilities found, severity levels.\n\n### 5. Acceptance Criteria Verification\n\nRead `docs\u002Fplan\u002F*\u002Fspec.md` and check each acceptance criterion:\n\nFor each `- [ ]` criterion in spec.md:\n1. Search codebase for evidence it was implemented.\n2. Check if related tests exist.\n3. Mark as verified or flag as missing.\n\n**Update spec.md checkboxes.** After verifying each criterion, use Edit tool to change `- [ ]` to `- [x]` in spec.md. Leaving verified criteria unchecked causes staleness across pipeline runs — check them off as you go.\n\n```\nAcceptance Criteria:\n  - [x] User can sign up with email — found in app\u002Fauth\u002Fsignup\u002Fpage.tsx + test\n  - [x] Dashboard shows project list — found in app\u002Fdashboard\u002Fpage.tsx\n  - [ ] Stripe checkout works — route exists but no test coverage\n```\n\nAfter updating checkboxes, commit: `git add docs\u002Fplan\u002F*\u002Fspec.md && git commit -m \"docs: update spec checkboxes (verified by review)\"`\n\n### 6. Code Quality Spot Check\n\nRead 3-5 key files (entry points, API routes, main components):\n- Check for TODO\u002FFIXME\u002FHACK comments that should be resolved\n- Check for console.log\u002Fprint statements left in production code\n- Check for proper error handling (try\u002Fcatch, error boundaries)\n- Check for proper loading\u002Ferror states in UI components\n\nReport specific file:line references for any issues found.\n\n### 7. Plan Completion Check\n\nRead `docs\u002Fplan\u002F*\u002Fplan.md`:\n- Count completed tasks `[x]` vs total tasks\n- Flag any `[ ]` or `[~]` tasks still remaining\n- Verify all phase checkpoints have SHAs\n\n### 8. Production Logs (if deployed)\n\nIf the project has been deployed (deploy URL in CLAUDE.md, or `.solo\u002Fstates\u002Fdeploy` exists if pipeline state directory is present), **check production logs for runtime errors**.\n\nRead the `logs` field from the stack YAML (`templates\u002Fstacks\u002F{stack}.yaml`) to get platform-specific commands.\n\n**Vercel (Next.js):**\n```bash\nvercel logs --output=short 2>&1 | tail -50\n```\nLook for: `Error`, `FUNCTION_INVOCATION_FAILED`, `504`, unhandled rejections, hydration mismatches.\n\n**Cloudflare Workers:**\n```bash\nwrangler tail --format=pretty 2>&1 | head -50\n```\nLook for: uncaught exceptions, D1 errors, R2 access failures.\n\n**Fly.io (Python API):**\n```bash\nfly logs --app {name} 2>&1 | tail -50\n```\nLook for: `ERROR`, `CRITICAL`, OOM, connection refused, unhealthy instances.\n\n**Supabase Edge Functions:**\n```bash\nsupabase functions logs --scroll 2>&1 | tail -30\n```\n\n**iOS (TestFlight):**\n- Check App Store Connect → TestFlight → Crashes\n- If local device: `log stream --predicate 'subsystem == \"com.{org}.{name}\"'`\n\n**Android:**\n```bash\nadb logcat '*:E' --format=time 2>&1 | tail -30\n```\n- Check Google Play Console → Android vitals → Crashes & ANRs\n\n**If no deploy yet:** skip this dimension, note in report as \"N\u002FA — not deployed\".\n\n**If logs show errors:**\n- Classify: startup crash vs runtime error vs intermittent\n- Add as FIX FIRST issues in the report\n- Include exact log lines as evidence\n\nReport:\n- Log source checked (platform, command used)\n- Errors found: count + severity\n- Error patterns (recurring vs one-off)\n- Status: CLEAN \u002F WARN \u002F ERRORS\n\n### 9. Dev Principles Compliance\n\nCheck adherence to dev principles. Look for `templates\u002Fprinciples\u002Fdev-principles.md` (bundled with this skill), or check CLAUDE.md or project docs for architecture and coding conventions.\n\nRead the dev principles file, then spot-check 3-5 key source files for violations:\n\n**SOLID:**\n- **SRP** — any god-class\u002Fgod-module doing auth + profile + email + notifications? Flag bloated files (>300 LOC with mixed responsibilities).\n- **DIP** — are services injected or hardcoded? Look for `new ConcreteService()` inside business logic instead of dependency injection.\n\n**DRY vs Rule of Three:**\n- Search for duplicated logic blocks (Grep for identical function signatures across files).\n- But don't flag 2-3 similar lines — duplication is OK until a pattern emerges.\n\n**KISS:**\n- Over-engineered abstractions for one-time operations?\n- Feature flags or backward-compat shims where a simple change would do?\n- Helpers\u002Futilities used only once?\n\n**Schemas-First (SGR):**\n- Are Pydantic\u002FZod schemas defined before logic? Or is raw data passed around?\n- Are API responses typed (not `any` \u002F `dict`)?\n- Validation at boundaries (user input, external APIs)?\n\n**Clean Architecture:**\n- Do dependencies point inward? Business logic should not import from UI\u002Fframework layer.\n- Is business logic framework-independent?\n\n**Error Handling:**\n- Fail-fast on invalid inputs? Or silent swallowing of errors?\n- User-facing errors are friendly? Internal errors have stack traces?\n\nReport:\n- Principles followed: list key ones observed\n- Violations found: with file:line references\n- Severity: MINOR (style) \u002F MAJOR (architecture) \u002F CRITICAL (data loss risk)\n\n### 10. Commit Quality\n\nCheck git history for the current track\u002Ffeature:\n\n```bash\ngit log --oneline --since=\"1 week ago\" 2>&1 | head -30\n```\n\n**Conventional commits format:**\n- Each commit follows `\u003Ctype>(\u003Cscope>): \u003Cdescription>` pattern\n- Types: `feat`, `fix`, `refactor`, `test`, `docs`, `chore`, `perf`, `style`\n- Flag: generic messages (\"fix\", \"update\", \"wip\", \"changes\"), missing type prefix, too-long titles (>72 chars)\n\n**Atomicity:**\n- Each commit = one logical change? Or monster commits with 20 files across unrelated features?\n- Revert-friendly? Could you `git revert` a single commit without side effects?\n\n**SHAs in plan.md:**\n- Check that completed tasks have `\u003C!-- sha:abc1234 -->` comments\n- Check that phase checkpoints have `\u003C!-- checkpoint:abc1234 -->`\n\n```bash\ngrep -c \"sha:\" docs\u002Fplan\u002F*\u002Fplan.md 2>\u002Fdev\u002Fnull || echo \"No SHAs found\"\n```\n\n**Pre-commit hooks:**\n\nRead the stack YAML `pre_commit` field to know what system is expected (husky\u002Fpre-commit\u002Flefthook) and what it should run (linter + formatter + type-checker). Then verify:\n\n```bash\n# Detect what's configured\n[ -f .husky\u002Fpre-commit ] && echo \"husky\" || [ -f .pre-commit-config.yaml ] && echo \"pre-commit\" || [ -f lefthook.yml ] && echo \"lefthook\" || echo \"none\"\n```\n\n- **Hooks installed?** Check config files exist AND hooks are wired (`core.hooksPath` for husky, `.git\u002Fhooks\u002Fpre-commit` for pre-commit\u002Flefthook).\n- **Hooks match stack?** Compare detected system with stack YAML `pre_commit` field. Flag mismatch.\n- **`--no-verify` bypasses?** Check if recent commits show signs of skipped hooks (e.g., lint violations that should've been caught). Flag as WARN.\n- **Not configured?** Flag as WARN recommendation — stack YAML expects `{pre_commit}` but nothing found.\n\nReport:\n- Total commits: {N}\n- Conventional format: {N}\u002F{M} compliant\n- Atomic commits: YES \u002F NO (with examples of violations)\n- Plan SHAs: {N}\u002F{M} tasks have SHAs\n- Pre-commit hooks: {ACTIVE \u002F NOT INSTALLED \u002F NOT CONFIGURED} (expected: {stack pre_commit})\n\n### 11. Documentation Freshness\n\nCheck that project documentation is up-to-date with the code.\n\n**Required files check:**\n```bash\nls -la CLAUDE.md README.md docs\u002Fprd.md docs\u002Fworkflow.md 2>&1\n```\n\n**CLAUDE.md:**\n- Does it reflect current tech stack, commands, directory structure?\n- Are recently added features\u002Fendpoints documented?\n- Grep for outdated references (old package names, removed files):\n  ```bash\n  # Check that files mentioned in CLAUDE.md actually exist\n  grep -oP '`[a-zA-Z0-9_.\u002F-]+\\.(ts|py|swift|kt|md)`' CLAUDE.md | while read f; do [ ! -f \"$f\" ] && echo \"MISSING: $f\"; done\n  ```\n\n**README.md:**\n- Does it have setup\u002Frun\u002Ftest\u002Fdeploy instructions?\n- Are the commands actually runnable?\n\n**docs\u002Fprd.md:**\n- Do features match what was actually built?\n- Are metrics and success criteria defined?\n\n**AICODE- comments:**\n```bash\ngrep -rn \"AICODE-TODO\" src\u002F app\u002F lib\u002F 2>\u002Fdev\u002Fnull | head -10\ngrep -rn \"AICODE-ASK\" src\u002F app\u002F lib\u002F 2>\u002Fdev\u002Fnull | head -10\n```\n- Flag unresolved `AICODE-TODO` items that were completed but not cleaned up\n- Flag unanswered `AICODE-ASK` questions\n- Check for `AICODE-NOTE` on complex\u002Fnon-obvious logic\n\n**Dead code check:**\n- Unused imports (linter should catch, but verify)\n- Orphaned files not imported anywhere\n- If `knip` available (Next.js): `pnpm knip 2>&1 | head -30`\n\nReport:\n- CLAUDE.md: CURRENT \u002F STALE \u002F MISSING\n- README.md: CURRENT \u002F STALE \u002F MISSING\n- docs\u002Fprd.md: CURRENT \u002F STALE \u002F MISSING\n- docs\u002Fworkflow.md: CURRENT \u002F STALE \u002F MISSING\n- AICODE-TODO unresolved: {N}\n- AICODE-ASK unanswered: {N}\n- Dead code: {files\u002Fexports found}\n\n### 12. Visual\u002FE2E Testing\n\nIf browser tools or device tools are available, run a visual smoke test.\n\n**Web projects (Playwright MCP or browser tools):**\n1. Start dev server (use `dev_server.command` from stack YAML, e.g. `pnpm dev`)\n2. Use Playwright MCP tools (or browser-use skill) to navigate to the main page\n3. Verify it loads without console errors, hydration mismatches, or React errors\n4. Navigate to 2-3 key pages (based on spec.md features)\n5. Take screenshots at desktop (1280px) and mobile (375px) viewports\n6. Look for broken images, missing styles, layout overflow\n\n**iOS projects (simulator):**\n1. Build for simulator: `xcodebuild -scheme {Name} -sdk iphonesimulator build`\n2. Install and launch on booted simulator\n3. Take screenshot of main screen\n4. Check simulator logs for crashes or assertion failures\n\n**Android projects (emulator):**\n1. Build debug APK: `.\u002Fgradlew assembleDebug`\n2. Install and launch on emulator\n3. Take screenshot of main activity\n4. Check logcat for crashes or ANRs: `adb logcat '*:E' --format=time -d 2>&1 | tail -20`\n\n**If tools are not available:** skip this dimension, note as \"N\u002FA — no browser\u002Fdevice tools\" in the report. Visual testing is never a blocker for SHIP verdict on its own.\n\nReport:\n- Platform tested: {browser \u002F simulator \u002F emulator \u002F N\u002FA}\n- Pages\u002Fscreens checked: {N}\n- Console errors: {N}\n- Visual issues: {NONE \u002F list}\n- Responsive: {PASS \u002F issues found}\n- Status: {PASS \u002F WARN \u002F FAIL \u002F N\u002FA}\n\n## Review Report\n\nGenerate the final report:\n\n```\nCode Review: {project-name}\nDate: {YYYY-MM-DD}\n\n## Verdict: {SHIP \u002F FIX FIRST \u002F BLOCK}\n\n### Summary\n{1-2 sentence overall assessment}\n\n### Tests\n- Total: {N} | Pass: {N} | Fail: {N} | Skip: {N}\n- Coverage: {N}%\n- Status: {PASS \u002F FAIL}\n\n### Linter\n- Errors: {N} | Warnings: {N}\n- Status: {PASS \u002F WARN \u002F FAIL}\n\n### Build\n- Status: {PASS \u002F FAIL}\n- Warnings: {N}\n\n### Security\n- Vulnerabilities: {N} (critical: {N}, high: {N}, moderate: {N})\n- Hardcoded secrets: {NONE \u002F FOUND}\n- Status: {PASS \u002F WARN \u002F FAIL}\n\n### Acceptance Criteria\n- Verified: {N}\u002F{M}\n- Missing: {list}\n- Status: {PASS \u002F PARTIAL \u002F FAIL}\n\n### Plan Progress\n- Tasks: {N}\u002F{M} complete\n- Phases: {N}\u002F{M} complete\n- Status: {COMPLETE \u002F IN PROGRESS}\n\n### Production Logs\n- Platform: {Vercel \u002F Cloudflare \u002F Fly.io \u002F N\u002FA}\n- Errors: {N} | Warnings: {N}\n- Status: {CLEAN \u002F WARN \u002F ERRORS \u002F N\u002FA}\n\n### Dev Principles\n- SOLID: {PASS \u002F violations found}\n- Schemas-first: {YES \u002F raw data found}\n- Error handling: {PASS \u002F issues found}\n- Status: {PASS \u002F WARN \u002F FAIL}\n\n### Commits\n- Total: {N} | Conventional: {N}\u002F{M}\n- Atomic: {YES \u002F NO}\n- Plan SHAs: {N}\u002F{M}\n- Status: {PASS \u002F WARN \u002F FAIL}\n\n### Documentation\n- CLAUDE.md: {CURRENT \u002F STALE \u002F MISSING}\n- README.md: {CURRENT \u002F STALE \u002F MISSING}\n- AICODE-TODO unresolved: {N}\n- Dead code: {NONE \u002F found}\n- Status: {PASS \u002F WARN \u002F FAIL}\n\n### Visual Testing\n- Platform: {browser \u002F simulator \u002F emulator \u002F N\u002FA}\n- Pages\u002Fscreens: {N}\n- Console errors: {N}\n- Visual issues: {NONE \u002F list}\n- Status: {PASS \u002F WARN \u002F FAIL \u002F N\u002FA}\n\n### Issues Found\n1. [{severity}] {description} — {file:line}\n2. [{severity}] {description} — {file:line}\n\n### Recommendations\n- {actionable recommendation}\n- {actionable recommendation}\n```\n\n**Verdict logic:**\n- **SHIP**: All tests pass, no security issues, acceptance criteria met, build succeeds, production logs clean, docs current, commits atomic, no critical visual issues\n- **FIX FIRST**: Minor issues (warnings, partial criteria, low-severity vulns, intermittent log errors, stale docs, non-conventional commits, minor SOLID violations, minor visual issues like layout overflow) — list what to fix\n- **BLOCK**: Failing tests, security vulnerabilities, missing critical features, production crashes in logs, missing CLAUDE.md\u002FREADME.md, critical architecture violations, app crashes on launch (simulator\u002Femulator) — do not ship\n\n## Post-Verdict: CLAUDE.md Revision\n\nAfter the verdict report, revise the project's CLAUDE.md to keep it lean and useful for future agents.\n\n### Steps:\n\n1. **Read CLAUDE.md** and check size: `wc -c CLAUDE.md`\n2. **Add learnings from this review:**\n   - New Do\u002FDon't rules discovered during review\n   - Updated commands, workflows, or architecture decisions\n   - Fixed issues or gotchas worth remembering\n   - Stack\u002Fdependency changes (new packages, removed deps)\n3. **If over 40,000 characters — trim ruthlessly:**\n   - Collapse completed phase\u002Fmilestone histories into one line each\n   - Remove verbose explanations — keep terse, actionable notes\n   - Remove duplicate info (same thing explained in multiple sections)\n   - Remove historical migration notes, old debugging context\n   - Remove examples that are obvious from code or covered by skill\u002Fdoc files\n   - Remove outdated troubleshooting for resolved issues\n4. **Verify result ≤ 40,000 characters** — if still over, cut least actionable content\n5. **Write updated CLAUDE.md**, update \"Last updated\" date\n\n### Priority (keep → cut):\n1. **ALWAYS KEEP:** Tech stack, directory structure, Do\u002FDon't rules, common commands, architecture decisions\n2. **KEEP:** Workflow instructions, troubleshooting for active issues, key file references\n3. **CONDENSE:** Phase histories (one line each), detailed examples, tool\u002FMCP listings\n4. **CUT FIRST:** Historical notes, verbose explanations, duplicated content, resolved issues\n\n### Rules:\n- Never remove Do\u002FDon't sections — critical guardrails\n- Preserve overall section structure and ordering\n- Every line must earn its place: \"would a future agent need this to do their job?\"\n- Commit the update: `git add CLAUDE.md && git commit -m \"docs: revise CLAUDE.md (post-review)\"`\n\n## AFTER CLAUDE.md revision — output signal EXACTLY ONCE:\n\nOutput pipeline signal ONLY if pipeline state directory (`.solo\u002Fstates\u002F`) exists.\n\n**Output the signal tag ONCE and ONLY ONCE.** Do not repeat it. The pipeline detects the first occurrence.\n\n**If SHIP:** output this exact line (once):\n```\n\u003Csolo:done\u002F>\n```\n\n**If FIX FIRST or BLOCK:**\n1. Open plan.md and APPEND a new phase with fix tasks (one `- [ ] Task` per issue found)\n2. Change plan.md status from `[x] Complete` to `[~] In Progress`\n3. Commit: `git add docs\u002Fplan\u002F && git commit -m \"fix: add review fix tasks\"`\n4. Output this exact line (once):\n```\n\u003Csolo:redo\u002F>\n```\n\nThe pipeline reads these tags and handles all marker files automatically. You do NOT need to create or delete any marker files yourself.\n**Output the signal tag once — the pipeline detects the first occurrence.**\n\n## Error Handling\n\n### Tests won't run\n**Cause:** Missing dependencies or test config.\n**Fix:** Run `npm install` \u002F `uv sync`, check test config exists (jest.config, pytest.ini).\n\n### Linter not configured\n**Cause:** No linter config file found.\n**Fix:** Note as a recommendation in the report, not a blocker.\n\n### Build fails\n**Cause:** Type errors, import issues, missing env vars.\n**Fix:** Report specific errors. This is a BLOCK verdict — must fix before shipping.\n\n## Two-Stage Review Pattern\n\nWhen reviewing significant work, use two stages:\n\n**Stage 1 — Spec Compliance:**\n- Does the implementation match spec.md requirements?\n- Are all acceptance criteria actually met (not just claimed)?\n- Any deviations from the plan? If so, are they justified improvements or problems?\n\n**Stage 2 — Code Quality:**\n- Architecture patterns, error handling, type safety\n- Test coverage and test quality\n- Security and performance\n- Code organization and maintainability\n\n## Verification Gate\n\n**No verdict without fresh evidence.**\n\nBefore writing any verdict (SHIP\u002FFIX\u002FBLOCK):\n1. **Run** the actual test\u002Fbuild\u002Flint commands (not cached results).\n2. **Read** full output — exit codes, pass\u002Ffail counts, error messages.\n3. **Confirm** the output matches your claim.\n4. **Only then** write the verdict with evidence.\n\nNever write \"tests should pass\" — run them and show the output.\n\n## Rationalizations Catalog\n\n| Thought | Reality |\n|---------|---------|\n| \"Tests were passing earlier\" | Run them NOW. Code changed since then. |\n| \"It's just a warning\" | Warnings become bugs. Report them. |\n| \"The build worked locally\" | Check the platform too. Environment differences matter. |\n| \"Security scan is overkill\" | One missed secret = data breach. Always scan. |\n| \"Good enough to ship\" | Quantify \"good enough\". Show the numbers. |\n| \"I already checked this\" | Fresh evidence only. Stale checks are worthless. |\n\n## Critical Rules\n\n1. **Run all checks** — do not skip dimensions even if project seems simple.\n2. **Be specific** — always include file:line references for issues.\n3. **Verdict must be justified** — every SHIP\u002FFIX\u002FBLOCK needs evidence from actual commands.\n4. **Don't auto-fix code** — report issues and add fix tasks to plan.md. Let `\u002Fbuild` fix them. Review only modifies plan.md, never source code.\n5. **Check acceptance criteria** — spec.md is the source of truth for \"done\".\n6. **Security is non-negotiable** — any hardcoded secret = BLOCK.\n7. **Fresh evidence only** — run commands before making claims. Never rely on memory.\n",{"name":5,"license":15,"metadata":24,"description":7,"allowed-tools":29,"argument-hint":30},{"author":25,"version":26,"openclaw":27},"fortunto2","1.1.1",{"emoji":28},"🔎","Read, Grep, Bash, Glob, Write, Edit, mcp__solograph__session_search, mcp__solograph__project_code_search, mcp__solograph__codegraph_query, mcp__solograph__codegraph_explain, mcp__solograph__web_search, mcp__context7__resolve-library-id, mcp__context7__query-docs","[focus-area]","main","f9d3fe7cf9ac8f1ea63a0d96e98e1f510a3814db","skills\u002Ffortunto2\u002Fsolo-review",1,"https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fopen-claw-skills","https:\u002F\u002Fgithub.com\u002Fjavimosch","intrane.fr","Building robust, open-source infrastructure for the commons. AI-leveraged, self-hosted, built to last.","2026-06-07 03:34:58",{"source":8,"skillName":5,"commands":41,"prompt":51,"zipAvailable":52},[42,45,48],{"tool":43,"command":44},"npx","npx skills add javimosch\u002Fopen-claw-skills --skill solo-review",{"tool":46,"command":47},"bunx","bunx skills add javimosch\u002Fopen-claw-skills --skill solo-review",{"tool":49,"command":50},"pnpm","pnpm dlx skills add javimosch\u002Fopen-claw-skills --skill solo-review","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fopen-claw-skills\u002Fblob\u002Fmain\u002Fskills\u002Ffortunto2\u002Fsolo-review\u002FSKILL.md",true,[],[55,63,70,77,84,91],{"skillNo":56,"name":57,"nameZh":6,"description":58,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":59,"categories":60,"keywords":61,"publishedAt":62},"01M778GGXG2VQ0YF31W2MKGMQZ","crowd-prompting","A marketplace where AI agents improve prompts, system instructions, tool descriptions, and other text-based content with domain expertise from real-world operations — and earn tokens for valuable contributions.","https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fopen-claw-skills\u002Fblob\u002Fmain\u002Fskills\u002Fzzadrian\u002Fcrowd-prompting\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":64,"name":65,"nameZh":6,"description":66,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":67,"categories":68,"keywords":69,"publishedAt":62},"01M778G7BH8SBJ9FJ6C3G7CB1K","ceaser","Interact with the Ceaser privacy protocol on Base L2 using the ceaser-mcp MCP tools. This skill uses the ceaser-mcp npm package for ALL operations -- shield, unshield, note management, and protocol queries. All ceaser tool calls use CLI subcommands (npx -y ceaser-mcp \u003Csubcommand>). Ceaser is a privacy-preserving ETH wrapper using Noir\u002FUltraHonk zero-knowledge proofs on Base L2 (chain ID 8453).","https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fopen-claw-skills\u002Fblob\u002Fmain\u002Fskills\u002Fzyra-v21\u002Fceaser\u002FSKILL.md",[],[],{"skillNo":71,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":74,"categories":75,"keywords":76,"publishedAt":62},"01M778FYCJQZWFKKX5FYMK3S47","ceaser-send","Fully automated private ETH transfer via Ceaser Protocol on Base L2 using the ceaser-mcp MCP tools. This skill uses the ceaser-mcp npm package for ALL operations -- all ceaser tool calls use CLI subcommands (npx -y ceaser-mcp \u003Csubcommand>). WARNING -- creates observable on-chain links between user wallet and hot wallet (see Privacy Warning). Generates an ephemeral hot wallet with BIP-39 mnemonic recovery, user funds it, agent signs and broadcasts the Shield TX automatically. Exactly one manual step required (funding the hot wallet with ETH). Uses Noir\u002FUltraHonk zero-knowledge proofs.","https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fopen-claw-skills\u002Fblob\u002Fmain\u002Fskills\u002Fzyra-v21\u002Fceaser-send\u002FSKILL.md",[],[],{"skillNo":78,"name":79,"nameZh":6,"description":80,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":81,"categories":82,"keywords":83,"publishedAt":62},"01M778FQQNSEYA3HJKZBPQ4M9Z","dilemm-ai","Compete in DilemmAI, the prisoner's dilemma AI arena at dilemm.ai. Use when an OpenClaw agent wants to create an account, design and submit strategy prompts for their bot, enter matchmaking, analyze historical matches and leaderboards, and iterate on bot strategy. Requires headless browser automation (no API available yet).","https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fopen-claw-skills\u002Fblob\u002Fmain\u002Fskills\u002Fzyori\u002Fdilemm-ai\u002FSKILL.md",[],[],{"skillNo":85,"name":86,"nameZh":6,"description":87,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":90,"publishedAt":62},"01M778FGVZX7QA2TF9BFMM37FY","moldium","Post and manage content on the Moldium blog platform. Triggered by \"post to Moldium\", \"write a blog post\", \"publish an article\", etc.","https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fopen-claw-skills\u002Fblob\u002Fmain\u002Fskills\u002Fzyom45\u002Fmoldium\u002FSKILL.md",[],[],{"skillNo":92,"name":93,"nameZh":6,"description":94,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":95,"categories":96,"keywords":97,"publishedAt":62},"01M778F88D8PQ9MQ0KNW2TJ8ZH","arxiv-paper-reviews","Interact with arXiv Crawler API to fetch papers, read reviews, submit comments, search papers, and import papers. Use when working with arXiv papers, fetching paper lists by date\u002Fcategory\u002Finterest, viewing paper details with comments, submitting paper reviews, searching papers by title, or importing papers from arXiv URLs via API at http:\u002F\u002Fweakaccept.top:8000\u002F.","https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fopen-claw-skills\u002Fblob\u002Fmain\u002Fskills\u002Fzxrys\u002Farxiv-paper-reviews\u002FSKILL.md",[],[]]