Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Auto-suggest optimal role and approach. Trigger with "review this", "check code quality", "best approach?", "how to improve?".
| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 48% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -20% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 205% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 84% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 173% | 0% |
A command that analyzes the current situation and automatically suggests the optimal role and approach.
bash/smart-review # Analyze current directory /smart-review <file/directory> # Analyze specific target
package.json, *.tsx, *.jsx, *.css, *.scss → frontendDockerfile, docker-compose.yml, *.yaml → architect*.test.js, *.spec.ts, test/, __tests__/ → qa*.rs, Cargo.toml, performance/ → performanceauth.js, security.yml, .env, config/auth/ → securitylogin.tsx, signup.js, jwt.js → security + frontendapi/auth/, middleware/auth/ → security + architectmobile/ + *.swift, *.kt, react-native/ → mobilewebpack.config.js, vite.config.js, large-dataset/ → performancecomponents/ + responsive.css → frontend + mobileapi/ + auth/ → security + architecterror.log, crash.log → analyzermemory leak, high CPU, slow query → performance + analyzerSQL injection, XSS, CSRF → security + analyzerbash$ /smart-review src/auth/login.js → "Authentication file detected" → "Analysis with security role recommended" → "Execute? [y]es / [n]o / [m]ore options"
bash$ /smart-review src/mobile/components/ → "📱🎨 Mobile + Frontend elements detected" → "Recommended approaches:" → "[1] mobile role alone" → "[2] frontend role alone" → "[3] multi-role mobile,frontend" → "[4] role-debate mobile,frontend"
bash$ /smart-review error.log → "⚠️ Error log detected" → "Starting root cause analysis with analyzer role" → "[Auto-execute] /role analyzer" $ /smart-review slow-api.log → "🐌 Performance issue detected" → "Recommended: [1]/role performance [2]/role-debate performance,analyzer"
bash$ /smart-review architecture-design.md → "🏗️🔒⚡ Architecture + Security + Performance elements detected" → "For complex design decisions, debate format recommended" → "[Recommended] /role-debate architect,security,performance" → "[Alternative] /multi-role architect,security,performance"
bash# Analyze current directory /smart-review "Suggest the optimal role and approach" # Analyze specific file /smart-review src/auth/login.js "Suggest the best review method for this file" # Analyze error log /smart-review error.log "Suggest the best approach to resolve this error"
bash$ /smart-review → "📊 Analyzing project..." → "React + TypeScript project detected" → "Authentication functionality + API + mobile support confirmed" → "" → "💡 Recommended workflow:" → "1. Check authentication with security" → "2. Evaluate UI/UX with frontend" → "3. Confirm mobile optimization with mobile" → "4. Review overall design with architect" → "" → "Auto-execute? [y]es / [s]elect role / [c]ustom"
bash$ /smart-review "How to set JWT expiration time" → "🤔 Technical design decision detected" → "This issue requires multiple expert perspectives" → "" → "Recommended approach:" → "/role-debate security,performance,frontend" → "Reason: Balance between security, performance, and UX is important"
bash# Analysis combined with file content cat src/auth/middleware.js /smart-review "Analyze this file from a security perspective" # Analysis combined with errors npm run build 2>&1 | tee build-error.log /smart-review build-error.log "Suggest ways to resolve build errors" # Design consultation /smart-review "Discuss whether to choose React Native or Progressive Web App"
Other measured skills in the registry, with their headline benchmark lift.