Analysis updated 2026-05-18
Run a private, self-hosted AI assistant for a household or small team
Chat with different specialist agent personas depending on the task
Connect the assistant to messaging apps like Feishu, DingTalk, or Discord
Delegate coding tasks to tools like Claude Code or Codex through the Agent Client Protocol
| tencentcloud/octop | cslawyer1985/claude-for-legal-zh | hero8152/infinite-canvas | |
|---|---|---|---|
| Stars | 265 | 264 | 266 |
| Language | Python | Python | Python |
| Setup difficulty | moderate | moderate | moderate |
| Complexity | 3/5 | 3/5 | 3/5 |
| Audience | developer | general | vibe coder |
Figures from each repo's GitHub metadata at analysis time.
One-line installer provisions Python automatically, extra setup is needed for IM channels and optional browser automation.
Octop is a self-hosted AI assistant that runs on your own computer instead of a company's servers, built for households and small teams rather than a single person. Instead of one general chatbot, it gives each user their own set of specialized agents they can switch between depending on the task, and everything, including conversations, files, and login credentials, stays on your machine in a single folder. You can talk to Octop through a web dashboard, a command line interface, or through messaging apps like Feishu, DingTalk, QQ, Discord, and WeCom, and it also exposes a plain HTTP and streaming API for developers who want to build on top of it. One nice touch is a library of sixteen personality templates based on the well known MBTI framework, so different agents can be given distinct characters depending on what you want them to help with, such as a work assistant versus a household planner. Security features are built in rather than added on: multiple users are kept separate, tool use requires approval, shell commands are restricted, and personal information can be automatically redacted. Storage is flexible too, supporting local disk, Docker containers, or cloud object storage, and an agent's memory can travel along with its workspace using a built-in memory system with full text search. For developers, Octop also plugs into the Agent Client Protocol, letting other coding tools use an Octop agent, or letting Octop delegate coding tasks out to tools like Claude Code, Codex, or OpenCode. It is built in Python with FastAPI on the backend and React with TypeScript on the frontend. Getting started is meant to be simple: a one-line installer script sets up an isolated Python environment automatically on macOS, Linux, or Windows, without needing Python pre-installed. The project is released under the MIT license, a permissive license that allows free use, modification, and commercial reuse as long as the copyright notice is kept.
A self-hosted, multi-user AI assistant with switchable specialist agents, IM channel integrations, and built-in privacy and security controls.
Mainly Python. The stack also includes Python, FastAPI, React.
MIT license: free to use, modify, and reuse commercially as long as you keep the copyright notice.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Verify against the repo before relying on details.