Deploy a team of AI agents that work together to run your business.
Website · Docs · Discord · Twitter
YokeBot is an open-source AI agent workforce platform. Create teams of specialized AI agents that collaborate autonomously — handling tasks, meetings, research, writing, analysis, and more — while you stay in control.
Think of it as hiring a team of AI employees that actually work together, share context through a knowledge base, and report back to you.
- 40+ Pre-Built Agents — Sales, marketing, engineering, HR, finance, legal, ops, and more — ready to deploy in seconds
- Voice Meetings — Real-time voice collaboration where agents speak (TTS) and you speak back (STT)
- Knowledge Base — Upload documents and agents use them as context via semantic search
- Task Management — Assign tasks to agents, track progress, set deadlines
- Workflows — Chain agents together into automated multi-step pipelines
- Data Tables — Structured data views with sorting, filtering, and export
- API Keys — Programmatic access to your agents and data
- Real-Time Collaboration — Raise your hand to cut the speaking queue during live meetings
- Self-Hosted — Run the entire platform on your own infrastructure
- Bring Your Own Keys — Use your own LLM API keys or let us handle it
The fastest way to get started:
- Sign up at yokebot.com
- Create a team and pick from 40+ pre-built agents
- Start a meeting or assign your first task
You get free starter credits to try everything out — no credit card required.
Run YokeBot on your own infrastructure with Docker:
# Clone the repo
git clone https://github.com/yokebots/yokebot.git
cd yokebot
# Configure environment
cp .env.example .env
# Edit .env with your API keys and database password
# Start everything
docker compose up -dThis starts the engine, dashboard, and a Postgres database with pgvector. Open http://localhost:3000 to access the dashboard.
See the Self-Hosted Setup Guide for detailed configuration and environment variables.
yokebot/
├── packages/
│ ├── engine/ # Agent runtime — Express 5, TypeScript
│ └── dashboard/ # Web UI — React 19, Vite, Tailwind CSS v4
├── skills/ # First-party agent skills
├── ee/ # Enterprise features
├── supabase/ # Database migrations
└── docker-compose.yml # Self-hosted deployment
Tech stack:
| Layer | Technology |
|---|---|
| Frontend | React 19, TypeScript, Vite, Tailwind CSS v4 |
| Backend | Express 5, TypeScript, Node.js |
| Database | PostgreSQL 17 + pgvector |
| Auth | Supabase (Google + GitHub OAuth) |
| Monorepo | pnpm workspaces |
YokeBot exposes a REST API for programmatic access. Generate an API key from your team settings, then:
curl https://api.yokebot.com/v1/agents \
-H "Authorization: Bearer yk_your_api_key"See the full API Reference for all endpoints.
Full documentation is available at yokebot.com/docs:
We welcome contributions! See CONTRIBUTING.md for:
- Development environment setup
- Project structure overview
- How to submit pull requests
- Core (
packages/engine,packages/dashboard,skills/) — AGPLv3 - Enterprise (
ee/) — YokeBot Enterprise License - SDKs & client libraries — MIT
- GitHub Discussions — Questions & ideas
- Discord — Chat with the team
- Twitter — Updates & announcements
- Issues — Bug reports & feature requests