> ## Documentation Index
> Fetch the complete documentation index at: https://docs.belvedir.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Python SDK

> Trace Python agents with the fractal-loop package.

Python agents use `fractal-loop` — the same concepts as the Node SDK, with context managers instead of callbacks. It auto-instruments the installed `anthropic` / `openai` / LangChain / LlamaIndex packages.

## 1. Install the SDK

```bash theme={null}
pip install fractal-loop
```

## 2. Initialize at process startup

Call `initialize()` once, before your agent starts making LLM calls. Unlike Next.js there is no bundler in the way, so no `instrumentModules` equivalent is needed:

```python theme={null}
import os
import fractal_loop as loop

loop.initialize(
    api_key=os.environ["FRACTAL_API_KEY"],
    app_name="my-agent",
)
```

## 3. Group agent work into sessions

Same as the Node SDK: `session` links every span under one session id (required for Tasks & Groups), and `task()` optionally marks task boundaries. Both are context managers and work inside `async def` bodies:

```python theme={null}
with loop.session(session_id=chat_id, user_id=user.id):
    with loop.task("send_email"):
        agent.run("email danny the report")
```

## 4. Flush before short-lived processes exit

Spans export in batches; scripts and serverless handlers can exit before the batch ships. Call `loop.flush()` before returning, or pass `disable_batch=True` to `initialize()` for local testing.