cockpit
llms.txt4 starsChanged 7 days agoMITView on GitHub
cockpit is the local environment coding agents work from on a product of one or many repositories. One reviewed cockpit.toml gives Claude Code, Codex CLI and OpenCode the same instructions, skills, permissions and guard hook (cockpit sync writes them, cockpit check fails on drift). cockpit session runs the product's own Compose stack on the live checkouts with a leased port family per checkout and starts the agent inside it. Work orders are graded by a fresh read-only agent run by…
In this repository4 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- verify-harness-capabilitySkill
- adopt-cockpitSkill
The file
Files
# cockpit > cockpit is the local environment coding agents work from on a product of one or many repositories. One reviewed `cockpit.toml` gives Claude Code, Codex CLI and OpenCode the same instructions, skills, permissions and guard hook (`cockpit sync` writes them, `cockpit check` fails on drift). `cockpit session` runs the product's own Compose stack on the live checkouts with a leased port family per checkout and starts the agent inside it. Work orders are graded by a fresh read-only agent run by default, or by a subagent of the working agent, the checks alone, the agent's own report, or nothing; the record names which. The instructions and guard need only Git and Node.js; the stack needs Docker Compose. Not a sandbox. Pre-1.0, macOS and Linux. ## Set cockpit up - [Getting started](https://raw.githubusercontent.com/b0dea/cockpit/main/docs/getting-started.md): the on-ramp, the safety contract and exact non-interactive commands for an agent setting cockpit up, and the bring-up sequence - [README](https://raw.githubusercontent.com/b0dea/cockpit/main/README.md): what cockpit is, the quickstart, a sample `cockpit.toml`, and its limits ## Reference - [CLI reference](https://raw.githubusercontent.com/b0dea/cockpit/main/docs/cli-reference.md): every command, flag, exit code and diagnostic - [Configuration essentials](https://raw.githubusercontent.com/b0dea/cockpit/main/docs/configuration-essentials.md): the `cockpit.toml` tables a typical cockpit sets, with minimal snippets - [Configuration reference](https://raw.githubusercontent.com/b0dea/cockpit/main/docs/configuration-reference.md): every `cockpit.toml` field with its type, default and constraints, and a link to the JSON Schema - [Concepts](https://raw.githubusercontent.com/b0dea/cockpit/main/docs/concepts.md): the terms the commands and diagnostics use ## Contributing - [AGENTS.md](https://raw.githubusercontent.com/b0dea/cockpit/main/AGENTS.md): the working guide for agents changing cockpit itself
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
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.