ct

Discover shell aliases and generate Claude Code documentation for tool replacements

Use when conducting comprehensive research - applies scientific methodology with adaptive strategies, multi-hop reasoning, and evidence-based synthesis

Use this agent proactively to guide refactoring decisions during code improvement and reactively to assess refactoring opportunities after tests pass (TDD's third step). Invoke when tests are green, when considering abstractions, or when reviewing code quality.

Simplifies and refines code for clarity, consistency, and maintainability while preserving all functionality. Focuses on recently modified code unless instructed otherwise. Language-agnostic.

Structured research specialist for external knowledge gathering

Fixes critical code review findings. Receives review findings, applies targeted fixes, and verifies tests pass. Used by the fix-loop skill.

Autonomous code review agent. Use proactively after code changes to analyse for best practices, security, performance, and potential issues. Use when the user asks for a code review.

Discover edge cases and test coverage gaps through systematic analysis. Use when analysing test coverage, finding bugs, hunting for edge cases, reviewing code for robustness, or when code-reviewer identifies test gaps. Also use when the user says things like "what could go wrong", "is this well tested", "find holes in my tests", "what am I missing", or asks about edge cases for any function or module.

Use when the user mentions duplicated code, copy-paste, code clones, repeated code, similar code across files, overlapping implementations, code quality audit, code smells, technical debt involving duplication, or refactoring preparation. Also for cross-language duplication checks, CI duplication warnings, and codebase redundancy scans. DO NOT TRIGGER for implementing refactoring (use incremental-refactoring), fixing bugs, writing tests, adding features, or code review.

Iterative review-fix cycle that eliminates all critical issues. Runs code-reviewer, fixes critical findings, verifies tests pass, and repeats until clean or max iterations reached. Use when the user says "review and fix", "find and fix bugs", "clean up the code", "fix all issues", "review then fix", or any request that combines finding problems with resolving them automatically.

Use when the user asks to refactor, clean up, restructure, extract methods, reduce complexity, or improve code structure without changing behavior. Triggers on: extract method, move function, simplify conditional, reduce nesting, split class. To FIND duplicates first, use duplicate-code-detector then return here. Not for features, bugs, or performance.

Use when optimizing backend, API, database, or system performance. Also use when response times are slow, queries need tuning, throughput is degraded, or someone says "just add caching." Triggers on: optimize API, backend slow, API latency, database performance, query optimization, server response time, connection pooling, N+1 queries, cache strategy, load testing. For frontend/UI performance (Core Web Vitals, Lighthouse accessibility), use frontend-production-quality instead.

Use when implementing auth, file uploads, payments, webhooks, OAuth/SSO, APIs, CI/CD, or external integrations. Also use when the user asks about security, security review, security audit, hardening, or vulnerability assessment for a feature or system. Applies STRIDE framework with DFD trust boundaries and DREAD scoring. Triggers on authentication, file upload, payment, multi-tenant, external API, webhook, OAuth, SSO, CI/CD, API gateway, WebSocket, background jobs, "is this secure", "security concerns", "security review", "make this secure", "harden this", "check security". Use this for ANY feature that handles user data, accepts external input, or crosses trust boundaries - even if it seems simple. If thinking I know security - use this anyway.

3 hooks across 3 events