ai-harness-doctorlisted

# ai-harness-doctor One-line purpose: run a repository's AI harness configuration through Checkup -> Treat -> Follow-up -> Efficacy, consolidating scattered or drifting agent instructions into the single source of truth `AGENTS.md` and keeping them guarded over time. ## When to use - The repository contains multiple rule files, such as `AGENTS.md`, `CLAUDE.md`, `.cursorrules`, `.cursor/rules/*.mdc`, `.windsurfrules`, `.github/copilot-instructions.md`, `GEMINI.md`, `.clinerules`, `.amazonq/rules/*.md`, `.junie/guidelines.md`, `WARP.md`, `.idx/airules.md`, `.goosehints`, `.kiro/steering/*.md`, `.augment-guidelines`, or `.kilocode/rules/*.md`. - The user asks to migrate, unify, or organize agent configs; run an agent config checkup; check agent config drift; consolidate agent configs; or audit an AI harness. - Tool-specific configs need to be downgraded to minimal stubs pointing at `AGENTS.md`, with CI or pre-commit drift guards added afterward. ## When not to use - Do not initialize an empty repository from scratch; that overlaps with the official `/init` flow. - Do not perform a full harness-platform refactor, such as converting scripts into a CLI, adding validation gates, or redesigning documentation layers; those belong to v2. - Do not generate language- or framework-specific standards from scratch, such as React, Go, or Python rules; those belong to v2 or a project-specific effort. - Do not adjudicate semantic conflicts for the user. Scripts perform deterministic mecha