agentleFS
HeyEddi-comHeyEddi-com/heyeddi-skills/AGENTS.md

heyeddi-skills

AGENTS.md0 starsChanged 7 days agoMITView on GitHub

Thin gates for every Agent chat in this project. Full playbooks and scripts stay in skills — load them when a gate says so (or via /skill). @heyeddi-intake, @heyeddi-product, @taste-penpot, @heyeddi-design, visual/UX auditors, @heyeddi-pr-review, @heyeddi-pr-respond, scaffolders / handoff. Managed by @heyeddi sync / auto-sync. Edit outside this block freely; content between the markers is replaced on sync. <!-- heyeddi-always-on:end -->

The file

Files
AGENTS.md

Agent instructions

<!-- heyeddi-always-on:start -->

HeyEddi always-on

Thin gates for every Agent chat in this project. Full playbooks and scripts stay in skills — load them when a gate says so (or via /skill).

Checklist

  1. Router / .heyeddi/ — Session start or ambiguous task: @heyeddi. Keep .heyeddi/skills-index current; follow .heyeddi/docs/intake/skill-routing.json when present. Do not freestyle the pipeline. Do not re-scaffold workflow every message.
  2. Setup — Before git / CI / commit / push / agent autonomy: @heyeddi-setup → verify_setup --check. Incomplete prefs fail those actions.
  3. Local-first — Before apply or other live cloud writes: @heyeddi-local-first → classify_command.py --check. Exit 1 → do not run. terraform plan is allowed and must be summarized.
  4. Engineering — Before feature code: @engineering-excellence plan gate. After edits: audit_engineering --check (errors fail; warns advisory).
  5. Tests (this project) — Not strict TDD. When you fix a bug: add or update a test that would have failed, run it, then fix. A pure frontend bug is a Playwright test: sign in with the project's test login and navigate the flow on the local app. If local cannot reproduce, ask before a test env, and ask again before production. On those environments: look and navigate only. No destructive action whatsoever. When you add behavior: ship a focused test with the change. Skip only for pure docs/typos or unautomatable cases (say so). Run the project's test command for what you touched. Detail: @heyeddi reference/reproduce-first.md.
  6. Prose — Any prose you author (chat, books, docs, UI, PRs): follow skill PROSE_ANTI_SLOP. Run verify_prose --check on touched surfaces; add --glob for non-default paths. No "not for books" carve-out.
  7. Ship — Before merge-ready claims: @pre-merge-gate --signoff (exit 0 required).

On demand (not every chat)

@heyeddi-intake, @heyeddi-product, @taste-penpot, @heyeddi-design, visual/UX auditors, @heyeddi-pr-review, @heyeddi-pr-respond, scaffolders / handoff.

Managed by @heyeddi sync / auto-sync. Edit outside this block freely; content between the markers is replaced on sync.

<!-- heyeddi-always-on:end -->

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.