Loop directory

Collect external links from docs weekly, archive dead ones via Wayback Machine, open a PR with replacements and a report of unfixed links.

Work through the fix plan, completing one task per loop: fix bugs, harden security, add tests, eliminate any, refactor duplicates, improve frontend.

Run post-merge scans on recent main merges, update docs, and propose minor fixes—escalate architectural changes.

Run every 15 minutes to classify CI failures, fix clear regressions in a worktree with verification, and escalate infrastructure or security test failures.

Run ci-triage on failing CI, fix clear regressions in isolation, verify before push, and update state—up to 3 attempts per cluster.

Work through a checkpoint-driven progress.md plan step-by-step, verify each task, and block if you hit budget limits or exhaustion.

Run an agent to identify and fix flaky tests until your CI pipeline passes completely.

Read the implementation plan, run the next phase and gate, tick progress, codex loop, open a PR, and write close-out—never merge.

Run a check loop every 5 minutes until your PR is merge-ready, optionally shipping when passing.

Execute planned project phases in sequence, audit milestones, close gaps, and push changes until the roadmap is complete.

Extract NumberInput and Toggle into shared components, stopping if Settings tests or goldens fail.

Every 5m, scan open PRs for CI failures or review comments; propose minimal fixes on allowlisted ones via worktree, escalating after 3 attempts.

Run dependency audit every 6 hours, auto-patch safely, escalate risky updates to dependency-sweeper-state.md.

Implement one approved story from your backlog, pass all checks, and commit with the owner's identity when done.

Scan the codebase for all multi-value error tuples, centralize domain structs into types.go, and replace with strongly-typed Result wrappers until all functions return single objects with nil-safe pointer receivers.

Execute the next unchecked step from plan.md with fast gates after each edit, then trigger post-implementation loop and open a PR without merging.

Execute each step in your plan, compile, pass all tests, commit, and open a PR.

Shepherd a multi-phase implementation plan through CI, decision gates, and code review until merged to trunk or delivery confirmed.

Optimize checkout-service code until the p95 latency benchmark hits 120ms and correctness tests pass, recording each iteration's change and result.

Narrate the root cause, diff summary, and regression test path until the test passes.

Verify Adobe/Lightroom docs, run targeted and full tests, get adversarial code reviews, and stop when bracket smoke passes or blockers are documented.

Replace all v1Client calls with v2Client, verify tests pass and TypeScript compiles clean, stopping after 20 turns.

Resolve every ESLint violation in the gen-ai frontend until linting passes cleanly.

Run the issue-triage skill every 2 hours, read current state, suggest labels for new items without auto-applying them, and escalate ambiguous cases.

Re-read the playbook each hour, fix operational drift, probe owners for liveness, and stand down any lane lacking side effects.

Pull items from a prioritized work queue and implement each spec item to completion, refilling the queue from spec Status sections until all are done.

Loop every 30m checking CI status and unresolved comments, fixing scoped issues until the PR is mergeable or a stop rule fires.

Run post-merge checks on recent main merges, update tracking docs, flag refactors for escalation.

Every 6 hours, scan dependencies for updates and CVEs, apply patches, verify in worktree, and escalate majors to your log.

Work through incomplete stories from prd.json, implement and test each, mark passes=true when done, repeat until all stories ship.