airweave

Open-source context retrieval layer for AI agents and RAG systems.

### What is Airweave? Airweave connects to your apps, tools, and databases, continuously syncs their data, and exposes it through a unified, LLM-friendly search interface. AI agents query Airweave to retrieve relevant, grounded, up-to-date context from multiple sources in a single request. ### Where it fits Airweave sits between your data sources and AI systems as shared retrieval infrastructure. It handles authentication, ingestion, syncing, indexing, and retrieval so you don't have to rebuild fragile pipelines for every agent or integration. ### How it works 1. **Connect** your apps, databases, and documents (50+ integrations) 2. **Airweave** syncs, indexes, and exposes your data through a unified retrieval layer 3. **Agents query** Airweave via our SDKs, REST API, MCP, or native integrations with popular agent frameworks 4. **Agents retrieve** relevant, grounded context on demand ## Quickstart ### Cloud-hosted: [app.airweave.ai](https://app.airweave.ai) ### Self-hosted ```bash git clone https://github.com/airweave-ai/airweave.git cd airweave ./start.sh ``` → http://localhost:8080 > Requires Docker and docker-compose ## Supported Integrations

## SDKs ```bash pip install airweave-sdk # Python npm install @airweave/sdk # TypeScript ``` ```python from airweave import AirweaveSDK client = AirweaveSDK(api_key="YOUR_API_KEY") results = client.collections.search.instant( readable_id="my-collection", query="Find recent failed payments" ) ``` ## CLI Search collections, manage sources, and trigger syncs from your terminal: ```bash pip install airweave-cli ``` ```bash airweave auth login airweave search "quarterly revenue figures" --collection finance-data ``` The CLI outputs rich interactive results in your terminal and clean JSON when piped — making it work for both developers and AI agents. ## Tech Stack - **Frontend**: [React/TypeScript](https://react.dev/) with [ShadCN](https://ui.shadcn.com/) - **Backend**: [FastAPI](https://fastapi.tiangolo.com/) (Python) - **Databases**: [PostgreSQL](https://www.postgresql.org/) (metadata), [Vespa](https://vespa.ai/) (vectors) - **Workers**: [Temporal](https://temporal.io/) (orchestration), [Redis](https://redis.io/) (pub/sub) - **Deployment**: [Docker Compose](https://docs.docker.com/compose/) (dev), [Kubernetes](https://kubernetes.io/) (prod) ## Contributing We welcome contributions! See our [Contributing Guide](CONTRIBUTING.md). ## License [MIT License](LICENSE)

Discord · Issues · Twitter