Agent Harness Skills

Use when designing quality snapshots, generated quality reports, structural metrics, debt thresholds, regression budgets, quality gates, or gradual cleanup loops.

Use when designing or refactoring AGENTS.md, CLAUDE.md, GEMINI.md, Cursor rules, GitHub instructions, source-of-truth navigation, or agent onboarding entrypoints.

Use when evaluating an existing repository's agent-readiness, harness maturity, validation surfaces, source-of-truth docs, evidence artifacts, or next smallest harness improvement.

Use when splitting changes into atomic commits, preparing commits from mixed worktrees, staging exact paths, including related task-state updates, writing Conventional Commits, or preventing unrelated changes.

Use when designing agent_chats or agents_chat records, delivery evidence summaries, linked tasks or commits, validation notes, risks, review notes, or handoff records.

Use when deciding how requirements should be captured in design docs, tasks.md, external task systems, exec plans, acceptance criteria, status updates, or planning source-of-truth files.

Use when designing repository validation commands, doctor scripts, test matrices, JSON or JUnit outputs, CI gates, smoke checks, or harness command surfaces.

Use when connecting observed behavior, logs, metrics, request IDs, run IDs, screenshots, traces, external dependency results, or artifacts into a runtime evidence loop.

Use when turning architecture, layering, directory ownership, dependency direction, file-size limits, choke points, baselines, or allowlists into repository checks.