tdd

Use when installing standard AI-agent project tooling such as AGENTS.md, PLAN.md, CONTEXT.md, and project-local skills.

Use when upgrading dependencies with changelog review, compatibility checks, and focused tests.

Use when updating project documentation to match current behavior, commands, and workflows.

Use when making Git changes while preserving user work and repository history.

Use when diagnosing failing CI jobs, logs, environment differences, and flaky checks.

Use when writing architecture decision records with context, decision, rationale, and consequences.

Use when preparing repository state, conventions, and next steps for another maintainer or agent.

Use when reviewing code for bugs, regressions, risks, and missing tests.

Use when a failure needs structured diagnosis before code changes.

Use when a technical claim should be challenged against primary documentation.

Use when improving architecture without broad rewrites or style-only churn.

Use when quickly building a working proof of concept with explicit tradeoffs.