A self-hosted personal AI assistant that runs on your own devices and answers you through chat apps like WhatsApp, Telegram, Slack, and Discord.
- Stars
- 4,444
- Forks
- 798
- Open issues
- 54
- Last commit
- 24 hours ago
What is it?
Rakazo is an open source AI agent platform for persistent bots that keep their own conversations, memory, routines and history. You can run it on your own machine or server with Docker, or use it from a web browser, an Electron desktop app, or the iPhone and Android apps. It is positioned as a self-hosted alternative to hosted assistants such as Grok Bot and Muse.
Bots work with your own model credentials through Pi and can use a sandboxed computer with browser, terminal, file and graphical desktop access. Sandboxes can run locally in Docker or on remote providers such as E2B, Daytona, CreateOS and Box. Integrations include Composio, Pipedream Connect, remote MCP servers and OpenAPI documents, and voice mode supports several third-party speech services.
The project is in beta. Installation uses a single installer script that pulls published Docker images, or a source checkout with Node.js and pnpm for development. The stack is built with TypeScript, React, Hono, PostgreSQL and Prisma.
Key features
- 01Persistent bots with their own memory, routines and conversation history
- 02Sandboxed computers with browser, terminal, file and desktop access
- 03Bring-your-own model credentials through Pi
- 04Voice mode with support for ElevenLabs, OpenAI, Cartesia and Fish Audio keys
- 05App integrations via Composio, Pipedream Connect, MCP and OpenAPI
- 06Available on web, Electron desktop, iPhone and Android
Use cases
- Self-hosting a personal AI assistant on your own machine or VPS
- Running bots that use your own model provider and keep data on your infrastructure
- Giving an AI agent isolated computers with browser and terminal access
- Experimenting with agent sandboxes across Docker and multiple cloud providers
Topics
FAQ
- What is rakazo?
- Self-hosted AI assistant and persistent bot platform where you choose the model and sandbox, available on web, desktop, iPhone and Android. Rakazo is an open source AI agent platform for persistent bots that keep their own conversations, memory, routines and history. You can run it on your own machine or server with Docker, or use it from a web browser, an Electron desktop app, or the iPhone and Android apps. It is positioned as a self-hosted alternative to hosted assistants such as Grok Bot and Muse.
- Is rakazo free to use? What license does it have?
- rakazo is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
- Is rakazo still maintained?
- Actively maintained. The last commit to rakazo was on Oct 17, 2026.
- What can I use rakazo for?
- Self-hosting a personal AI assistant on your own machine or VPS Running bots that use your own model provider and keep data on your infrastructure Giving an AI agent isolated computers with browser and terminal access Experimenting with agent sandboxes across Docker and multiple cloud providers
- Where can I get rakazo?
- The source code and installation instructions are on GitHub: https://github.com/elie222/rakazo
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
A self-hosted web app that converts files between over a thousand formats using tools like FFmpeg, ImageMagick, Pandoc and LibreOffice.
Open-source, self-hostable backend platform for building apps, offering Auth, Databases, Storage, Functions, Messaging, Realtime, Hosting and MCP access for agents.
An open-source framework for building universal native apps with React that run on Android, iOS, and the web.
A self-hosted Docker dashboard that aggregates dozens of public OSINT feeds (aircraft, ships, satellites, earthquakes, CCTV, radio) onto one live map with an optional AI agent interface.
A software development methodology and composable skills library that guides coding agents through brainstorming, planning, TDD, and subagent-driven implementation.