agentleFS
Sign inStart free
NeogenuityNeogenuity/harness-kit/llms.txt

harness-kit

llms.txt0 starsChanged 4 months agoMITView on GitHub

Harness Engineering for reliable coding agents: canonical context, in-turn feedback, guardrails, executable verification, and continuous improvement. A Claude Code skill/plugin that makes this environment portable across Claude Code, Cursor, Codex, and OpenCode, vendoring it so the target repo depends on nothing at runtime. The repository root is itself a live installation of the kit (dogfooding); only the plugins/harness-kit/ directory ships to users.

The file

Create a free account to mount
Files
llms.txt
# harness-kit

> Harness Engineering for reliable coding agents: canonical context, in-turn
> feedback, guardrails, executable verification, and continuous improvement.
> A Claude Code skill/plugin that makes this environment portable across Claude
> Code, Cursor, Codex, and OpenCode, vendoring it so the target repo depends on
> nothing at runtime.

The repository root is itself a live installation of the kit (dogfooding);
only the `plugins/harness-kit/` directory ships to users.

## Docs

- [README](README.md): value proposition, install, layout, status
- [AGENTS.md](AGENTS.md): table of contents of this repo's own knowledge base
- [The harness pattern](plugins/harness-kit/skills/harness-kit/references/pattern.md): design principles, anatomy, the three layers, upgrade model
- [Provider matrix](plugins/harness-kit/skills/harness-kit/references/provider-matrix.md): per-harness file locations, hook events, payloads — key facts carry verification stamps, with a Sources section to re-check against
- [Architecture](ARCHITECTURE.md): how this repo embodies the pattern
- [Decision records](docs/architecture/decisions/README.md): why advisory stop-hooks, vendoring, generated stubs, verification stamps, manifest self-protection

## Usage

- [The skill](plugins/harness-kit/skills/harness-kit/SKILL.md): init / audit / add-skill / add-agent / add-hook / update workflows
- [Contributing](CONTRIBUTING.md): quality gates, template discipline
- [Changelog](CHANGELOG.md)

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free