Self-hostable AI second brain that answers questions from the web and your own documents, with custom agents and scheduled automations.
yc-software/qm
A self-hosted multiplayer AI agent for teams, usable in Slack and on the web, with per-person and per-room isolated workspaces.
- Stars
- 15,650
- Forks
- 1,945
- Open issues
- 1,020
- Last commit
- 3 hours ago
What is it?
QM is an agent harness designed for teams rather than individuals. Each employee gets an isolated workspace with its own memory, files, credentials view, permissions, scheduled jobs, web apps and durable sandbox, and the agent can also work inside shared Slack channels, group messages and projects.
The core is headless and generic; company-specific configuration, custom tools, skills and infrastructure live in a separate deployment directory managed by the qm CLI. It supports Pi and Claude Code as interchangeable harnesses, and Pi can run models from many providers. Deployment targets include Fly and AWS, and the deployment runs in the operator's own cloud account with Postgres for durable sessions and state.
Security is configured per organization with three postures (Strict, Auto and Dangerous) and an independent sharing setting (Isolated or Open), plus optional content screening and a predeclared command policy.
Key features
- 01Per-person and per-room isolated memory, files, permissions and sandboxes.
- 02Works in both Slack and a web app with the same identity and configuration.
- 03Supports Pi and Claude Code as swappable agent harnesses and many model providers.
- 04Admin controls for org-wide security posture, sharing and available models.
- 05Shared skills with admin-gated promotion and import from git repositories.
- 06Crons, watches and inbound webhooks for background work.
Use cases
- Giving a startup team a shared AI assistant in Slack channels for project tracking and follow-ups.
- Searching internal notes, email, documents and databases together with web results.
- Building and publishing internal web apps for specific employees.
- Letting engineers use the agent inside a repository to run tests, open PRs and monitor CI.
Quick start
- 1
Initialize a deployment repository with the qm CLI
npm exec --yes --package=@yc-software/qm@latest -- qm init . --org <slug> --target <fly-or-aws> - 2
Install the deployment repository's dependencies
npm install - 3
Follow the generated deployment skill to set up infrastructure, sign-in, credentials, and deploy
Things to know
- Each deployment runs in your own cloud account and requires provider setup, so it is not a one-click install.
- Durable sessions need a Postgres database configured via DATABASE_URL and SESSION_STORE=postgres; otherwise sessions are lost on restart.
Featured in Repolode Weekly
- Issue #2 · A Rust headless browser for agents, and typed .sql imports
A self-hosted multiplayer AI agent for teams, used from Slack or the web with one identity. Each person and each room gets isolated memory, files, permissions and sandboxes, and the harness can be swapped between Pi and Claude Code.
In collections
Topics
FAQ
- What is qm?
- A self-hosted multiplayer AI agent for teams, usable in Slack and on the web, with per-person and per-room isolated workspaces. QM is an agent harness designed for teams rather than individuals. Each employee gets an isolated workspace with its own memory, files, credentials view, permissions, scheduled jobs, web apps and durable sandbox, and the agent can also work inside shared Slack channels, group messages and projects.
- Is qm free to use? What license does it have?
- qm is open source under the MIT license. Check the license terms before using it in commercial products.
- Is qm still maintained?
- Actively maintained. The last commit to qm was on Nov 7, 2026.
- What can I use qm for?
- Giving a startup team a shared AI assistant in Slack channels for project tracking and follow-ups. Searching internal notes, email, documents and databases together with web results. Building and publishing internal web apps for specific employees. Letting engineers use the agent inside a repository to run tests, open PRs and monitor CI.
- How do I install qm?
- Initialize a deployment repository with the qm CLI: npm exec --yes --package=@yc-software/qm@latest -- qm init . --org <slug> --target <fly-or-aws> Install the deployment repository's dependencies: npm install Follow the generated deployment skill to set up infrastructure, sign-in, credentials, and deploy
- Where can I get qm?
- The source code and installation instructions are on GitHub: https://github.com/yc-software/qm
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
A framework for training, evaluating and deploying harness adapters that evolve an external layer around an agent without changing its model or harness.
Open-source multi-agent AI assistant that runs tasks across web, desktop, files and chat apps like Telegram, Feishu and WeCom.
A Rust base for building vertical AI agents, with a CLI, 42+ LLM providers, 25+ chat channels and a modular kernel.
A filesystem-first TypeScript framework for building durable AI agents, with tools, skills, channels and schedules defined as files.
A self-hosted personal AI assistant that runs on your own devices and answers you through messaging apps like WhatsApp, Telegram, Slack and Discord.