me

Use when designing systems, architectures, or APIs, when multiple valid approaches exist with no single obvious answer, or when user explicitly requests parallel/competing solutions

Create PR — commit, push, PR, wait for merge.

Use when verifying a running application end-to-end through its real interface — a web UI, a CLI, or a TUI — by writing and executing agent-run "scenario cards" against a freshly built instance with falsifiable assertions. Trigger on "test it end to end", "prove the UI actually works", "write/run a scenario", or after a change touches a user-facing surface that unit tests can't fully cover. Not for unit tests, pure code review, or API-only checks.

Investigate a question against high-trust primary sources and capture the findings as a Markdown file in the repo. Use when the user wants a topic researched, docs or API facts gathered, or reading legwork delegated to a background agent.

Use when preparing to deploy to production, asked to "ship", "release", or "deploy", or when you need to verify a deploy succeeded or plan a rollback. Covers the full flow: pre-deploy checks, the deploy itself, and post-deploy verification with rollback on failure.

Use when asked to "verify this", "qa", "does this implementation work?", or "test this feature". Verifies the current implementation in context and reports `PASS`, `PARTIAL`, or `FAIL` with evidence.

Use when creating or improving an agent skill, writing a SKILL.md or its description, fixing a skill that triggers too rarely or too often, or tuning one with SkillOpt or SkillOpt-Sleep from benchmark tasks or provider session history.

Use when writing, reviewing, or improving PRDs, product requirements documents, product specs, feature requirements, launch requirements, or product discovery writeups before product or engineering work begins.

Use when writing, reviewing, or improving RFCs, requests for comments, technical design docs, architecture proposals, engineering decision records, API proposals, migration plans, or cross-team technical decisions.