Skip to content

safe-write-pattern.md

Latest commit

 

History

History
38 lines (31 loc) · 1.52 KB

safe-write-pattern.md

File metadata and controls

38 lines (31 loc) · 1.52 KB
Raw

Safe Write Pattern (MCP + Loops)

Any time a loop needs to change external state (create PR, update ticket, post comment that looks official), follow this pattern:

  1. Discovery / Triage (read MCP or local git)
  2. Worktree for code changes (isolated)
  3. Implementer sub-agent (minimal-fix or equivalent)
  4. Verifier sub-agent (separate session or higher-reasoning model)
  5. Propose only:
    • Open a draft PR (or comment with diff link + summary)
    • Or create/update a ticket with "proposed by loop — human review requested"
    • Sign everything: 🤖 Loop Engineering — <Pattern Name>
  6. Record in state (what was proposed, link, timestamp, risk)
  7. Human gate (or very narrow allowlist + extra verifier for trivial patches)
  8. Prune / close stale proposals on next runs

Example MCP + Prompt Snippet (Grok)

/loop 15m 
Use github-propose MCP (read + draft PR + comment).
Run pr-review-triage on open PRs.
For low-risk, allowlisted items that pass verifier:
  - open worktree
  - minimal-fix
  - loop-verifier
  - post signed comment + link to worktree (or open draft PR)
Never merge. Update pr-babysitter-state.md.

Why This Works

  • The loop has agency (it can discover and propose quickly).
  • The human (or strict allowlist) retains final authority on mutation.
  • Full context lives in STATE + the PR/ticket, so comprehension debt stays manageable.
  • Easy to pause: scheduler_delete or disable the Action.

This is the recommended default for all L2+ loops that touch external systems.