#348 · Primary category: AI Coding Assistants
repository-harness
Turn any repo into an agent-ready workspace for Claude Code, Codex, Cursor, and other coding agents.
Project last updated:08/13/26
GitHub Stars
1.2K
Forks
429
Contributors
8
License
MIT
Why we included this project
Coding agents like Claude Code, Codex, and Cursor tend to flounder on real repositories when the intent behind a change lives only in chat and nothing in the repo points to the authoritative documents. This project installs a small protocol that makes a codebase legible to those agents: a compact AGENTS.md entrypoint, a navigable documentation map, durable plans only when a change actually spans sessions, and explicit judgment boundaries so an agent stops before inventing product policy. It stays out of the way for typo-level fixes but forces structure for long or ambiguous work, and its safe updater three-way merges managed files and stages conflicts rather than clobbering local edits. Teams that want consistent agent behavior across many repos will appreciate the install script and versioned harness binary, and the explicit-only onboarding and invariant skills keep control over when agents act.
Articles for this project
No articles for this project yet.
To suggest a topic or contribute an article, contact us.
Related projects in this category
superpowers
An agentic skills framework & software development methodology that works.
ECC
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
skills
Skills for Real Engineers. Straight from my .agents directory.
opencode
The open source coding agent.
andrej-karpathy-skills
A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.