Octop
A self-hosted AI assistant that runs multiple agents for several users on your own machine.
- Stars
- 11.5k
- Forks
- 1.5k
- Licence
- MIT
- Language
- Python
- Last update
- 4 days ago
- Type
- App
What it is
Octop is an open-source AI assistant that runs as a single process on your own computer. It serves a web dashboard, a command-line interface, chat integrations such as Feishu, Telegram and Discord, and scheduled tasks, with all data kept under ~/.octop/. Each user can have their own set of specialized agents, which suits households and small teams who want to keep their conversations and credentials local.
- Let a dedicated agent draft weekly reports, organize notes and manage a schedule
- Set up cron-style jobs in natural language so an agent runs or pushes tasks each day
- Ask an agent to automate web tasks with Browser AI+ (forms, screenshots)
- Ground answers in your own documents using the built-in knowledge base (RAG)
Great for
- Households or small teams that want a shared, self-hosted assistant
- People who want to keep conversations and credentials on their own machine
- Users who already chat through Feishu, DingTalk, WeCom, Telegram or Discord and want an AI assistant there
- Developers who want to delegate coding tasks to OpenCode or Claude Code through ACP
Not ideal for
- People who need a mature, fully stable product: the release is tagged 1.0.3b1 and AgentTeams is marked Beta in the README
- Users who want a zero-setup hosted service, since Octop is designed to run on your own machine
- Anyone unwilling to manage a self-hosted process, its database and its connectors
How to get it
- 1
Download the app
You'll need: Python 3.12+ (listed in the README for the source install)
Clone the repository
git clone https://github.com/TencentCloud/Octop - 2
Run it
- Download the release for your platform from the GitHub releases page, or clone the repository and install from source
- Run the first-run setup wizard with `octop init`
- Start Octop with `octop run`
- Open the web dashboard and create your admin account
- 3
Get the most out of it
- Enable the interactive API docs at /api/docs by setting "enable_api_docs": true in config.json (off by default)
- Connect IM channels such as Feishu or Telegram to chat with your agents from your phone or group chats
- Use the expert library and MBTI persona templates to give each agent a distinct role
- Add Connectors (OAuth and MCP) to extend what agents can access
- Choose a workspace backend (local disk, Docker sandbox, PostgreSQL or COS/S3) to fit your storage needs
Downloads
Strengths
- Keeps data local: the README states all data lives under ~/.octop/ with SQLite by default
- One process serves the web UI, CLI, chat channels and cron, which keeps setup simpler
- Supports multiple users and multiple experts per user, each with its own workspace
Trade-offs
- The README describes many features; some, like AgentTeams, are marked Beta
- Several features (remote desktop, browser automation, terminal AI) add setup and security considerations that the README only partly explains in the visible text
- The README shown here is truncated, so detailed setup and configuration steps could not be verified
How it compares
| Project | Stars | Type | Level | Updated | Best for |
|---|---|---|---|---|---|
| Octop (this) | 11.5k | App | intermediate | 4 days ago | Households or small teams that want a shared, self-hosted assistant |
| anything-llm | 67.8k | App | beginner | 7 days ago | People who want a private ChatGPT-style assistant over their own PDFs and text files |
| LibreChat | 46.6k | App | intermediate | 4 days ago | Teams that want one shared chat interface across several AI providers |
| invisible_dots | 32.4k | AI model / agent | intermediate | 6 days ago | Developers who want a self-hosted agent and are comfortable with a command-line setup |
| learn-harness-engineering | 22.9k | Learning resource | beginner | 15 days ago | Developers new to building or controlling AI coding agents |
| paperclip | 110k | App | advanced | 4 days ago | Developers already running several AI coding agents who want one place to track them |
Stats synced from GitHub. The assessment above was drafted with AI from the project's README on 2026-10-29 and may miss recent changes — check the README before relying on a command. MIT licence: free to use, modify and distribute, including commercially, as long as you keep the copyright notice.