sofias-sdk-lite

Build multi-node agents on a declarative graph and run them against RabbitMQ.

What you get

• An agent graph — LLM nodes, function nodes, delegation nodes (agent-to-agent), aggregator nodes (fan-in), and planner nodes (dynamic DAGs), connected with explicit routing strategies.
• A messaging layer — pydantic wire models (AgentTaskMessage, StreamFragment, delegation contracts) and a RabbitMQ transport built on aio-pika and RabbitMQ Streams.
• A runner — AgentRunner consumes tasks from a queue, resolves per-request settings, executes your agent, and streams the response back — with graceful shutdown, retries, and circuit breakers built in.
• An LLM that is already wired — LLM nodes talk to the Sofias gateway (or any OpenAI-compatible endpoint) using the model, URL and key from the agent's settings or the SOFIAS_LLM_* environment. create_llm() builds the client when you need it explicitly; LLMCallable stays a protocol for anything the bundled clients do not cover — see LLM integration.

Nothing in this package depends on a private backend or an internal config service. Where the model lives is deployment configuration, not agent code.

Where to go next

• Quickstart — build and execute your first agent in a few minutes, no broker required.
• Concepts → Agents and graphs — nodes, routing, contracts.
• Messaging → Running an agent — wire the same agent up to RabbitMQ with AgentRunner.
• Reference — full API reference.

Install

pip install sofias-sdk-lite

Requires Python 3.11+. A running RabbitMQ broker (with the Streams plugin) is only needed once you use AgentRunner or streaming responses — the agent graph itself has no broker dependency and can be built and executed directly, as in the quickstart.