Documented, reusable loop patterns that have been (or can be) run in real environments.
Each pattern answers:
- What problem it solves
- Recommended scheduling
- Required skills / state shape
- Verification approach
- Human hand-off strategy
- Tool-specific notes (Grok, Claude Code, Codex, OpenClaw, Opencode, GitHub Actions)
Machine-readable index: registry.yaml
- Pick a pattern: pattern-picker.md
- Scaffold with
npx @cobusgreyling/loop-init . --pattern <name> --tool grok(or--tool opencode/--tool claude) or copy fromstarters/ - Copy skills from
templates/if customizing beyond the starter - Set up scheduling (
/loop,scheduler_create, GitHub Action, Codex Automation) - Run week one in L1 report-only mode before enabling fixes
- Audit with
npx @cobusgreyling/loop-audit . --suggest
Use templates/pattern-template.md, add an entry to registry.yaml, and open a PR. See CONTRIBUTING.md.
