Metadata-Version: 2.4
Name: bigrag
Version: 2026.5.7
Summary: Python SDK for the bigRAG platform
Project-URL: Homepage, https://github.com/yoginth/bigrag
Project-URL: Documentation, https://bigrag.dev/docs/sdks/python
Project-URL: Repository, https://github.com/yoginth/bigrag
Project-URL: Issues, https://github.com/yoginth/bigrag/issues
Author-email: bigRAG <dev@bigrag.dev>
License-Expression: MIT
Keywords: bigrag,embeddings,rag,search,vector
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Typing :: Typed
Requires-Python: >=3.11
Requires-Dist: httpx>=0.28.0
Description-Content-Type: text/markdown

# bigRAG Python SDK

Async Python client for the bigRAG API.

```bash
pip install bigrag
```

```python
import asyncio

from bigrag import BigRAG


async def main() -> None:
    async with BigRAG(api_key="bigrag_sk_...", base_url="http://localhost:4000") as client:
        doc = await client.documents.upload("docs", "/path/to/paper.pdf")
        result = await client.queries.query("docs", {"query": "What is RAG?"})
        print(doc["id"], result["total"])


asyncio.run(main())
```

The SDK is fully typed, ships `py.typed`, and uses CalVer releases like `2026.5.7`.

## Namespaces

- `client.collections` for collection CRUD, stats, re-embedding, and event streams.
- `client.documents` for uploads, batch operations, file URLs, and status polling.
- `client.queries` for single, multi-collection, and batch retrieval queries.
- `client.vectors` for raw vector upsert and delete.
- `client.webhooks` for webhook management and delivery replay.
- `client.auth` for session login, setup, preferences, and identity.
- `client.admin` for users, API keys, access logs, audit logs, runtime settings, backups, admin realtime streams, connector config, embedding presets, and MCP server keys.
- `client.connectors.google` for Google Drive account, file browsing, sources, and sync jobs.
- `client.evaluations` for golden-set retrieval evaluations.

## Authentication

API-key endpoints accept `api_key` or the `BIGRAG_API_KEY` environment variable.
Session-only admin endpoints can be used after calling `client.auth.login(...)` or
`client.auth.setup(...)`; the client keeps the session cookie in its underlying
`httpx.AsyncClient`.

## Versioning

Published artifacts use CalVer in the form `YYYY.M.D`.
