Forks and stars mean people are trying this in their own stacks. If you run a loop from this repo (or adapted from it), add yourself here via PR.
Show your readiness level in your README:
npx @cobusgreyling/loop-audit . --badgePaste the markdown output into your README. Re-run after you graduate L1 → L2 → L3.
Fast path: open the Add Adopter issue — we'll add your row.
Or open a PR that adds a row to the table below:
| Field | What to include |
|---|---|
| Project | Repo link or product name |
| Pattern(s) | e.g. Daily Triage + Issue Triage |
| Tool | Grok, Claude Code, Codex, Cursor, Windsurf, GitHub Actions, mixed |
| Level | L1 / L2 / L3 (honest) |
| Notes | One line — what worked or what broke |
| Project | Pattern(s) | Tool | Level | Notes |
|---|---|---|---|---|
| loop-engineering (this repo) | Daily Triage, Changelog Drafter, audit dogfood | GitHub Actions + Grok | L3 | Reference implementation — loop-audit on every PR; readiness score 100 |
| loop-engineering (maintainer) | Post-Merge Cleanup | Grok + GitHub Actions | L1→L2 | Off-peak scan; verifier caught doc/API drift — see story |
| loop-engineering (maintainer) | Issue Triage + Daily Triage | Grok | L1 | Issue queue feeder: propose labels only week one; pairs with morning STATE.md triage |
| loop-engineering (maintainer) | PR Babysitter, Dependency Sweeper | Grok / Claude Code | L2 | Assisted fixes in worktrees; human gate on merges; patch-only deps |
Your project here — see CONTRIBUTING.md.
Prefer chat over a PR? Post in GitHub Discussions → Show your loop with:
- Which pattern you picked and why
- Your first
/loopor scheduler command - One surprise (good or bad)
Failure reports are first-class — see stories/.