Debug responsive layout by rendering at real mobile/tablet viewports and measuring, rather than guessing from a desktop screen. Use for any layout that breaks on smaller screens.

Use when a bug report arrives — turn it into a minimal failing test, confirm it fails for the right reason (RED), then fix (GREEN), then keep the test as a permanent regression guard.

Use when cutting a release — semver decision table (breaking/feature/fix → major/minor/patch), Keep a Changelog format, and the full release checklist from version bump to announce.

Write a production-quality README — what it does, quickstart, usage, config, and contributing — that gets a new user running in minutes. Use when a project lacks a README or its README is stale.

Use when reading Python, JavaScript, Go, or Java stack traces — read from bottom up, find YOUR code first, extract the reproduction from the trace.

Profile Python to find real hotspots (cProfile/py-spy) and fix the actual bottleneck instead of guessing. Use when code is too slow and you don't know where time goes.

Use when example-based tests aren't enough — test invariants with randomly generated inputs, shrinking to minimal counterexamples. Use Hypothesis (Python) or fast-check (JS).

Build a mental map of an unfamiliar repo before editing it — read AGENTS.md/README, structure, tests, and conventions. Use on first contact with a codebase.

Wire `pre-commit` to run linters, formatters, type-checks, and secret scans automatically on every commit. Use to catch issues at commit time instead of in CI.

Use when opening, reviewing, or managing a pull request — branch naming, draft→ready transition, PR description template, the review loop, and merge strategy selection.