Use when a question requires comprehensive evidence gathering from multiple sources, systematic synthesis, and traceable citations…
Use when removing old APIs, migrating to new patterns, or cleaning up legacy code — treats code as liability and follows a structured…
Use when a bug or performance issue is still fuzzy — build the fastest feedback loop first, rank the leading hypotheses, and instrument…
Use when code has changed and documentation may be out of date — identifies stale docs, finds gaps between implementation and docs, and…
Use when you need end-to-end tests for critical user flows that unit tests can't cover — scaffolds Playwright/Cypress tests from the happy…
Use when you need to evaluate an LLM pipeline or AI feature systematically — sets up an eval harness with test cases, scoring rubrics, and…
Use when implementation is complete and tests pass but the branch itself is not yet resolved — verify tests, detect worktree vs. normal…
Use when a plan must be stress-tested against the codebase's existing language and decisions — grill it against current docs, glossary…
Use when work is changing sessions, agents, or machines and the next pass needs a compact handoff document with current state, open…
Use when a sprint or feature is complete and ready to ship — tags the version, generates GitHub Release notes, runs rollout or smoke…
Use when a task must stay inside a narrow file or directory boundary, or when risky commands need an explicit stop rule — define the…
Use when the goal is practical vulnerability discovery for responsible disclosure or bounty submission — focuses on remotely reachable…
Use when you want a quick security pass on code changes or dependencies — checks OWASP Top 10 patterns, runs dependency audits, and…
Use when implementing framework-specific or library-specific code — verify non-trivial APIs against official documentation before writing…
Use when a bug is non-obvious or has resisted quick fixes — applies a 4-phase root cause analysis to find and permanently resolve the…
Use when starting a new feature or function to write failing tests first, then implement the minimal code to pass (Red→Green→Refactor)
Use when a single issue or external-contributor pull request needs structured triage — classify it, verify the claim if needed, request…
Use when you need multiple branches checked out at once — create isolated working directories for parallel development without cloning the…
Use when a page or component needs a structured UX audit — quick scan applies 6 Krug-inspired usability checks; deep audit covers 15…
Use before claiming any task is done — run the exact command that proves the fix works, read the output, and only then report success.
Review a pull request without posting any comments.