Openbuden

Openbuden is a multi-agent Discord system where each agent runs as its own bot, shares a common context, and can call tools through MCP. It is designed for teams that want collaborative agents that can reason together while staying specialized.

Highlights

• One Discord bot per agent
• Shared conversation context across all agents
• LLM-backed decision making via LiteLLM
• Tool system with permissions and MCP bridge
• Simple CLI to configure, test, and run

Requirements

• Python 3.14+
• A Discord account and server
• Discord bot tokens for each agent
• An LLM API key (OpenAI, Anthropic, or Groq)

Quickstart

1. Create and activate a virtual environment:

   uv venv .venv && source .venv/bin/activate

2. Install in editable mode:

   uv pip install -e .

3. Create base configuration:

   openbuden config

4. Add your first agent:

   openbuden add-agent

5. Start Openbuden:

   openbuden start

Configuration

Openbuden uses two config locations:

• ~/.config/openbuden/settings.yaml for shared defaults
• .env in the project root for OPENBUDEN_LLM_API_KEY

To create them, run:

openbuden config

Common Commands

• openbuden config — create or update the base settings
• openbuden add-agent — create a new agent
• openbuden list — list agents and their tools
• openbuden tools — show registered tools and permissions
• openbuden test — verify LLM connectivity
• openbuden start — start all Discord bots

Documentation

See GUIDE.md for full documentation, examples, and troubleshooting.

Project Structure

src/openbuden/
  agent/      # Agent logic and shared context
  config/     # YAML config loading and schemas
  discord/    # Discord bot runtime
  llm/        # LiteLLM client and prompt builder
  tools/      # Tool registry, MCP bridge, executor

License

TBD