Leon is an open-source personal AI assistant that runs locally and can use tools, memory and context to complete tasks.
ownpilot/OwnPilot
Self-hosted, privacy-first personal AI assistant platform with autonomous agents, 250+ tools, MCP support, and Telegram and WhatsApp connectivity.
Open-source alternative to:ChatGPT
- Stars
- 430
- Forks
- 64
- Open issues
- 5
- Last commit
- last month
What is it?
OwnPilot is a self-hosted personal AI assistant platform built with TypeScript, a Hono API gateway, a React web UI, and PostgreSQL storage. It lets you connect multiple AI providers, including OpenAI, Anthropic, Google, and many aggregators or local models such as Ollama, and keeps personal data such as notes, tasks, contacts, calendar, and expenses in your own database.
Beyond chat, it provides autonomous "Claw" agents with isolated workspaces, scheduled or event-driven execution, and subclaw orchestration, along with soul agents that have persistent identities and heartbeat cycles. Tools cover files, code execution, browsing, email, git, and more, and the platform can act as an MCP client and server. Messages can be handled through Telegram and WhatsApp, and IoT devices can be controlled over MQTT.
The project is MIT licensed and runs via Docker Compose or from source with Node.js 22+ and pnpm.
Key features
- 01Connects to many AI providers, including OpenAI, Anthropic, Google, and local models via Ollama or LM Studio.
- 02Runs autonomous Claw agents with isolated workspaces and single-shot, interval, or event-driven modes.
- 03Provides 250+ built-in tools for files, code, web, email, git, and personal data.
- 04Acts as both an MCP client and an MCP server.
- 05Handles conversations through Telegram and WhatsApp channels.
- 06Controls IoT edge devices over MQTT.
Use cases
- Running a private AI assistant that keeps notes, tasks, and contacts on your own server.
- Automating recurring research or monitoring jobs with scheduled autonomous agents.
- Chatting with an AI assistant through Telegram or WhatsApp while data stays self-hosted.
- Connecting local models and cloud providers to one interface with per-task model routing.
Things to know
- The project is at version 0.8.3, which is pre-1.0, so interfaces and behavior may change.
- It requires Node.js 22 or newer, and the README points to a separate setup guide for installation.
In collections
Topics
FAQ
- What is OwnPilot?
- Self-hosted, privacy-first personal AI assistant platform with autonomous agents, 250+ tools, MCP support, and Telegram and WhatsApp connectivity. OwnPilot is a self-hosted personal AI assistant platform built with TypeScript, a Hono API gateway, a React web UI, and PostgreSQL storage. It lets you connect multiple AI providers, including OpenAI, Anthropic, Google, and many aggregators or local models such as Ollama, and keeps personal data such as notes, tasks, contacts, calendar, and expenses in your own database.
- Is OwnPilot free to use? What license does it have?
- OwnPilot is open source under the MIT license. Check the license terms before using it in commercial products.
- Is OwnPilot still maintained?
- Updated within 6 months. The last commit to OwnPilot was on Sep 8, 2026.
- What can I use OwnPilot for?
- Running a private AI assistant that keeps notes, tasks, and contacts on your own server. Automating recurring research or monitoring jobs with scheduled autonomous agents. Chatting with an AI assistant through Telegram or WhatsApp while data stays self-hosted. Connecting local models and cloud providers to one interface with per-task model routing.
- Where can I get OwnPilot?
- The source code and installation instructions are on GitHub: https://github.com/ownpilot/OwnPilot
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
An open-source, OpenClaw-style AI assistant with persistent memory that runs on Android, iOS, desktop, and the web.
Self-hostable personal AI assistant as a single Go binary, combining LLM chat, RAG knowledge bases, and agent loops for browsing, shell, and MCP tools.
A self-hosted personal AI agent server in a single Rust binary, with sandboxed tool execution, multiple LLM providers, and chat channels like Telegram and Discord.
A self-hosted personal AI assistant that runs on your own devices and answers you through messaging apps like WhatsApp, Telegram, Slack and Discord.
A local-first, provider-neutral personal AI workspace with chat, reviewed memory, auditable tools, MCP support, and isolated project plugins.