Audit and build the infrastructure a repo needs so agents can work autonomously — boot scripts, smoke tests, CI/CD gates, dev environment setup, observability, worktree isolation, and .worktreeinclude handling for Codex or Claude worktrees. Use when a repo can't boot, tests are broken or missing, there's no dev environment, agents can't verify their work, worktrees miss ignored local config, or agents need human help to get anything done. Do not use for reviewing an existing diff or for documentation-only cleanup.
Run the bundled Codex/Claude autoreview helper as a structured second-model closeout for local changes, pull requests, branch diffs, or commits: choose the target, validate findings, rerun focused tests, and repeat until clean. Use when explicitly asked for autoreview, Codex/Claude review, or a tool-backed final review after implementation. Do not use for builder verification or an independent multi-agent ship decision.
Update repo documentation and durable agent-facing artifacts such as AGENTS.md, README.md, docs/, specs, plans, decisions, and runbooks. Use when code, skill, or infrastructure changes risk doc drift or when documentation needs cleanup or restructuring. Do not use for code review, runtime verification, or boot/readiness infrastructure setup.
Set up or align a repository's GitHub collaboration and delivery surface: repo settings, branch/ruleset policy, PR and security templates, Actions hardening, GitHub Environments, release workflows, and deploy workflows. Use when standardizing GitHub setup for repos, CI/CD, publishing versioned packages, or deploying running apps; route app deploy details to deploy references and package publish details to release references.
Ban direct `useEffect` in React code. Use when writing, refactoring, reviewing, or migrating React components or hooks that import, call, add, or replace direct `useEffect`; when an agent reaches for effects for derived state, fetching, event reactions, resets, or external sync; or when adding lint/agent rules for a no-direct-useEffect policy. Do not use for ordinary React work with no effect smell, non-React code, or legitimate effect architecture outside React.
Review an existing diff, branch, pull request, or codebase with at least two independent reviewer subagents, scale the reviewer set to the risk, then synthesize their evidence into a ship decision. Use when the user asks to review this PR, check this diff independently, decide whether someone else's change is safe to merge, triage PR risk, or follow up after runtime proof. Produces a `ship it` / `needs review` / `blocked` verdict. Do not use to self-check a change you just authored.
Audit, score, or improve existing skills with Tessl review output, metadata checks, repo conventions, and skill-authoring best practices. Use when creating or revising a skill, asking why a skill did not activate, checking skill quality, comparing a skill against `AGENTS.md`, `CLAUDE.md`, or repo rules, or preparing a skill for publish. Do not use to verify general application code or to rewrite unrelated docs.
Self-check an agent's own completed change before independent review — the builder-owned pre-review sanity pass. Use when running repo guardrails, exercising changed behavior end to end, collecting evidence, or catching obvious self-correctable issues. Produces a `ready for review` / `needs more work` / `blocked` verdict — never a ship decision. If the repo cannot be booted or exercised reliably, report `blocked` with the exact missing infrastructure and required setup. Do not use to audit someone else's diff, branch, or PR.