Loop directory

Work through your PRD's highest-priority failing user story, implement it, pass QA checks, and repeat until all stories pass.

Run linters on modified files and exit when zero violations remain across all changes.

Scan, plan, and refactor all boolean naming violations across the codebase until 100% green without stopping.

Run static checks and git diff until both pass or hit a blocker after 6 attempts.

Run wiki-linter on a cadence to catch and fix frontmatter drift, stopping when all checks pass.

Convert manual process notes into a stepwise agent workflow with tool calls, owners, and checkpoints until all gates pass.

Scan and refactor all regexp usage across the codebase to use thread-safe lazyregex.New() with structured diagnostics until all tests pass.

Run npm test on test/auth until it exits 0, with only src/auth files modified and no test changes.

Run a tick every hour, audit each agent against the playbook, replace stuck lanes, and push side effects until all delegated work is done.

Implement a requested change, run documented checks, and surface verification evidence in the transcript until complete or blocked after 20 turns.

Take story P2-2 from spec through implementation, Codex dev, and T11 code review until it commits green to fidelity-perception-arc-2026-06-19.

Execute the ship spec: write docs, pass all tests and macro checks, deploy or block with reasons, and deliver final acceptance criteria.

Refactor src/utils.js into focused modules under 200 lines each, keeping all exports and signatures intact, until npm run build and npm test both pass with matching test counts.

Run checkout tests repeatedly until 20 consecutive passes, logging failure modes and hypotheses after each iteration.

Your agent runs autonomously toward a goal and exits 0 on success or 2 if blocked, enabling CI integration.

Convert all src/ JS files to TypeScript with strict annotations until npm test passes cleanly.

Run a full review pass inline, apply validated blocking findings at their owning layer, re-run fresh until the gate clears or blockers plateau—capped at 3 rounds with escalation.

A second, read-only loop that babysits a first one. Every ten minutes it reads the run's state file and latest log, runs the test suite so its evidence is an exit code rather than a self-report, and applies four explicit escalation triggers (no progress across two checks, repeated identical stack trace, budget exceeded, merge-queue conflict). On any trigger it prints ESCALATE and stops so a human looks; it never restarts, edits, merges, or pushes. Exits when the state file reports PASS or DONE, and after twelve checks regardless. Adapted from ECC's loop-operator agent (affaan-m/ECC, MIT), which lists the same triggers but has no cadence, cap, check command, or exit.

A recurring six-phase check (build, types, lint, tests with coverage, secret and debug-log grep, diff size) that prints one fixed-format report every fifteen minutes so you catch drift before the PR, not in review. It is strictly read-only: the loop reports, you decide what to fix. It exits when the tests pass and the report reads READY twice running, escalates instead of retrying when the same check fails three times, and stops after eight checks no matter what. Adapted from ECC's verification-loop skill (affaan-m/ECC, MIT), whose 'continuous mode' names the cadence but has no turn cap, exit, or failure path; all three are added here.

Two reviewers from different model families, each in a fresh context, must both pass an objective rubric before a PR is opened — the maker-checker split with real model diversity. Each failed round fixes only what was flagged and re-reviews with reviewers that have no memory of the last round, so nothing anchors on prior findings. Hard cap of three rounds, tests may never be edited to pass, and the loop ends by opening a PR rather than pushing. Adapted from ECC's /santa-loop (affaan-m/ECC, MIT), which auto-pushes on agreement and has no verifiable exit; this version adds the test-exit condition and the human merge gate.

The generic persistent-task Ralph pattern: iterate on one named task until its acceptance criteria are met and the suite is green, bounded by a 15-turn cap and a no-progress stop. Anti-reward-hacking clauses keep it from editing tests to go green, and it hands off via a PR instead of merging. Swap the check command for your stack's.

Build and test a new /users/:id/avatar route until it passes tests, lint, and matches the OpenAPI schema.

Work through each task in .toh/plan.md, fix blockers, until the build command succeeds or you hit 40 turns.

Run pre-QA tests repeatedly, diagnose and fix failures, report only on blockers or green.

Shape and execute infrastructure readiness objectives by gathering evidence, seeking approval at decision boundaries, then validating provisioning, configuration, and rollback in non-production before production deployment.

Discover and execute a benchmark optimization goal, freeze the protocol, compare candidates under identical conditions, and deploy only validated improvements with regression protection.

Review diffs iteratively, validate findings, self-fix each one, and restart until zero critical/high/medium issues remain.

Work through every task in PLAN.md, pausing for credentials or destructive actions, until pnpm test and pnpm build both succeed.

Refactor all files to TypeScript and run tests until the full suite passes.

Run verification commands until all pass, preventing Check Agent from stopping prematurely.