owncast-agent¶

Owncast API + MCP Server + A2A Agent for the agent-utilities ecosystem — the typed, deterministic control surface for an Owncast self-hosted live-streaming and chat server.

Official documentation

This site is the canonical reference for owncast-agent, maintained alongside every release.

Overview¶

owncast-agent wraps the Owncast admin and public REST API with typed, deterministic MCP tools and an optional Pydantic-AI agent server. It provides:

OwncastApi — a requests-based REST client over the Owncast API, organized by domain (system, chat, followers, configuration, integrations, auth).
Action-routed MCP tools — four togglable tool groups (internal, objects, external, chat) that dispatch to the client by action name, keeping the LLM tool surface compact.
An A2A agent server — a Pydantic-AI agent (console script owncast-agent) that consumes the MCP tools, with an optional web interface and OpenTelemetry tracing.

Explore the documentation¶

Installation — pip, source, extras, and the prebuilt Docker image.
Deployment — run the MCP and agent servers, Docker Compose, Caddy + Technitium.
Usage — the MCP tools, the OwncastApi client, and the CLI.
Backing Platform — deploy Owncast with Docker.
Architecture — pipeline, layered client, agent topology.
Concepts — the CONCEPT:OC-* registry.

Quick start¶

pip install owncast-agent
owncast-mcp                       # stdio MCP server (default transport)

Connect it to an Owncast server:

export OWNCAST_URL=http://your-owncast:8080
export OWNCAST_TOKEN=your_owncast_token
owncast-mcp --transport streamable-http --host 0.0.0.0 --port 8000

See Installation and Deployment for the full matrix (PyPI extras, Docker image, all transports, the agent server, reverse proxy, DNS).