# AgentPod

> AgentPod is a fleet and facilities console for agent runtimes: the console for the machines your agents live on. It attaches to runtimes you already run, detecting Claude Code, Codex, OpenCode, Hermes, OpenClaw and Pi in place rather than asking you to migrate anything. Every node dials out to the hub over a WebSocket, so a laptop behind NAT works the same as a VPS. From one place you can read a workspace, tail a log, open a shell and check health on any station; turn on the bridge and each station gets a chat identity you can message, and managed skills can be verified as actually loaded. The hub is a Bun service with a Postgres database and the console is a static site, all self-hosted.

## Documentation Sets

- [Abridged documentation](https://docs.agentpod.dev/llms-small.txt): a compact version of the documentation for AgentPod, with non-essential content removed
- [Complete documentation](https://docs.agentpod.dev/llms-full.txt): the full documentation for AgentPod

## Notes

- The complete documentation includes all content from the official documentation
- The content is automatically generated from the same source as the official documentation

## Optional

- [AgentPod](https://agentpod.dev): The product site.
- [Source](https://github.com/SuperJackfruitLabs/agentpod): The AgentPod repository on GitHub.