Clawborate

Community Edition

Clawborate is a platform for creating and managing collaborative AI agent teams. Define agent roles, assign tasks through a coordinator, and watch your team work together — each agent has its own identity, memory, and specialization.

Architecture

flowchart TB
    Browser["Browser (:3000)"]
    Nginx["Nginx<br/>/ → Frontend · /api/ → Backend · /ow/ → Open WebUI"]
    Frontend["Frontend (Next.js)"]
    Backend["Backend (Python)"]
    OpenWebUI["Open WebUI (chat UI)"]
    Postgres["PostgreSQL"]

    subgraph Agents["Agent Containers (Docker)"]
        A1["Agent 1"]
        A2["Agent 2"]
        A3["Agent 3"]
        AN["Agent N"]
    end

    Browser --> Nginx
    Nginx --> Frontend
    Nginx --> Backend
    Nginx --> OpenWebUI
    Backend --> Postgres
    Backend --> Agents

Loading

Backend manages teams, spawns agent containers via Docker API, and bridges chat between users and agents.

Agent containers are independent runtime instances, each with its own LLM connection, identity, and workspace. Agents communicate through a shared notification bus.

Supported Agent Runtimes

• OpenClaw — general-purpose agent runtime
• Hermes — coming soon
• Claude — coming soon

Open WebUI provides the chat interface — users talk to agents through it, and agent replies are routed back through the backend.

Quick Start

Prerequisites

• Docker and Docker Compose
• An LLM API key (Anthropic, OpenAI, or any OpenAI-compatible endpoint)

Setup

# 1. Clone the repo
git clone https://github.com/clawborate/clawborate-community.git
cd clawborate-community

# 2. Create your environment file
cp .env.example .env

# 3. Edit .env — at minimum, set these:
#    APP_DB_PASSWORD   — database password (generate a random one)
#    JWT_SECRET        — auth token secret (generate a random one)
#    LLM_BASE_URL      — your LLM API endpoint
#    LLM_API_KEY       — your LLM API key
#    LLM_MODEL         — model name (e.g. claude-sonnet-4-20250514)

# 4. Start everything
docker compose up -d

Generate secure passwords:

python3 -c "import secrets; print(secrets.token_hex(16))"   # APP_DB_PASSWORD
python3 -c "import secrets; print(secrets.token_hex(32))"   # JWT_SECRET

First Login

Open http://localhost:3000 and log in with the default credentials:

• Email: test@test.com
• Password: test

(Change these in .env via DEFAULT_USER_EMAIL / DEFAULT_USER_PASSWORD before first run.)

Create a Team

1. Go to Teams and click Create Team
2. Select the ACE Development Team template (4 agents: coordinator, developer, designer, tester)
3. Wait for all agent containers to start (health checks turn green)
4. Go to Settings > LLM and configure your API key, then click Save and Apply
5. Start chatting with your team through the coordinator

Configuration

All configuration is in .env. See .env.example for the full list of options.

LLM Providers

Provider	Variables to set
Anthropic	LLM_PROVIDER=anthropic, ANTHROPIC_API_KEY
OpenAI	LLM_PROVIDER=openai, OPENAI_API_KEY
Ollama (local)	LLM_PROVIDER=ollama, OLLAMA_BASE_URL=http://host.docker.internal:11434
Custom	LLM_PROVIDER=custom, LLM_BASE_URL, LLM_API_KEY, LLM_MODEL

LLM settings can also be configured after startup via the Settings page in the UI.

Customizing Agents

Agent definitions live in agent-arch/openclaw/. Each agent has an identity, personality, and memory that you can edit. See agent-arch/README.md for details.

Managing the Stack

# View logs
docker compose logs -f backend

# Restart
docker compose restart

# Stop
docker compose down

# Full reset (wipes all data)
docker compose down -v

Contributing

Contributions are welcome. See CONTRIBUTING.md for setup, scope, and how to open a pull request.

License

Community edition. See LICENSE for details.