This repository also configures its own agents. See what agents tells them →
Skill Claude CodeCodex
Coordinates bounded subagent review, intent-preserving fixes, and validation after a meaningful implementation slice; use only when the user explicitly asks to run Garfield or its review-fix-verify loop before handoff, not for standalone reviews, CI iteration, brainstorming, or skill authoring.
Skill Claude CodeCodex
Use when transcript, commit, Sentry, GitHub, CI, review, or Garfield history suggests recurring failure patterns that might become hard deterministic checks. Analyze the evidence, reject judgment-based ideas, and output only mechanically enforceable lint, schema, CI, workflow, or deterministic test rules. Do not use…
Skill Claude CodeCodex
Designs, prototypes, reviews, and polishes intentional, accessible visual artifacts with HTML, CSS, SVG, and JavaScript. Use for UI design, design systems, wireframes, HTML decks, interactive prototypes, visual variations, accessibility audits, and design-quality reviews.
Skill Claude CodeCodex
Designs and reviews tests for agents and LLM-backed workflows. Use when coverage involves model behavior, tool choice, semantic output, flaky evals, mocked agent paths, brittle assertions, or fixture overfitting; do not use for ordinary deterministic tests or general prompt experiments.
Skill Codex
Triages a Gmail inbox using private per-account policy, recipient and relationship evidence, audited spam/archive actions, verified unsubscribe, and corrections. Use for inbox cleanup, scheduled triage, or user corrections; not for drafting replies, deleting mail, or Gmail filters.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: