Self-hosted personal AI agent framework in Python with a browser WebUI, terminal client, chat app channels, tools, memory and MCP support.
CherryHQ/stella
Self-hosted, multi-user AI agent platform where teams share domain-specific agents reachable from Telegram, Discord, Feishu and other chat tools.
- Stars
- 53
- Forks
- 14
- Open issues
- 161
- Last commit
- yesterday
What is it?
Stella lets a domain owner configure an AI agent once, with its own instructions, skills, tools, knowledge and memory rules, and then lets the rest of a team ask it for help in the chat apps they already use. Teammates do not need to learn the underlying finance, HR or internal systems to get work done.
It runs as a single-tenant, multi-user, multi-agent deployment. Each user keeps a separate memory with each agent, and each agent has its own role, model, skills, tools, schedules and sandboxed workspace. Users bring their own model API keys and can reach agents from Telegram, Discord, QQ, Feishu, DingTalk, WeChat, the Web UI or a terminal TUI, with memory shared across channels.
The project is written in Go and distributed as a single binary, with a Web UI served on port 25678. It is marked as under heavy development, so APIs and config formats may change, and it is not recommended for production yet.
Key features
- 01Shared agents with per-user, per-agent memory
- 02Connects to Telegram, Discord, QQ, Feishu, DingTalk, WeChat, Web UI and terminal
- 03Per-agent model, skills, tools and schedules
- 04Sandboxed workspaces with controlled tool access
- 05MCP server integration over HTTP or SSE
- 06Self-hosted, using your own model API keys
Use cases
- Letting a finance or HR expert's repeated questions be answered by a shared agent
- Running a team assistant in Feishu or DingTalk for back-office tasks
- Setting up recurring reminders and digests that notify the right people
- Hosting a personal AI assistant on your own server across Telegram and the terminal
Topics
FAQ
- What is stella?
- Self-hosted, multi-user AI agent platform where teams share domain-specific agents reachable from Telegram, Discord, Feishu and other chat tools. Stella lets a domain owner configure an AI agent once, with its own instructions, skills, tools, knowledge and memory rules, and then lets the rest of a team ask it for help in the chat apps they already use. Teammates do not need to learn the underlying finance, HR or internal systems to get work done.
- Is stella free to use? What license does it have?
- stella is open source under the AGPL-3.0 license. Check the license terms before using it in commercial products.
- Is stella still maintained?
- Actively maintained. The last commit to stella was on Oct 21, 2026.
- What can I use stella for?
- Letting a finance or HR expert's repeated questions be answered by a shared agent Running a team assistant in Feishu or DingTalk for back-office tasks Setting up recurring reminders and digests that notify the right people Hosting a personal AI assistant on your own server across Telegram and the terminal
- Where can I get stella?
- The source code and installation instructions are on GitHub: https://github.com/CherryHQ/stella
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Open-source personal AI assistant and Agent Harness that plans tasks, runs tools and skills, and keeps memory across web and IM channels.
A self-hosted personal AI assistant that runs on your machine or in the cloud and connects to chat apps like DingTalk, Lark, Discord and Telegram.
Self-hosted Java/Spring Boot AI assistant platform with multi-user workspaces, multi-provider model failover, an LLM Wiki, MCP and Skills, and IM channel integrations.
A lightweight personal AI assistant written in Go that runs on low-cost hardware and connects to many LLM providers and chat apps.
A self-hosted personal AI agent platform where a crew of isolated agents shares long-term memory and reaches you through terminal, WhatsApp, Telegram, Slack and more.