Report Abuse

Basic Information

Airweave is a platform designed to let agents search data across applications and stores by indexing and vectorizing content into a semantically searchable knowledge base. The repository provides a complete system that connects to apps, productivity tools, databases, and document stores, extracts and transforms entities and content, creates embeddings, and exposes a standardized search interface for agents via a REST API or an MCP endpoint. It includes a dashboard frontend, backend services, SDKs for Python and TypeScript, and local deployment scripts using Docker Compose. The platform handles authentication, incremental syncs, versioning, multi-tenant configuration, and serving vectors through a vector database, enabling developers and builders to create or host semantic search services that agents can query across many integrated sources.

Links

Categorization

App Details

Features
Airweave ships connectors for 25+ integrations covering productivity apps, databases, document stores, and developer platforms. It provides automated data synchronization and incremental updates using content hashing, an entity extraction and transformation pipeline, semantic search powered by vector embeddings, and versioning for data changes. The architecture supports multi-tenant deployments with OAuth2, white-labeling for SaaS builders, and exposes functionality via a REST API and MCP server semantics. The stack documented in the repo includes a React/TypeScript frontend, FastAPI backend, PostgreSQL for metadata, Qdrant for vector storage, and deployment tooling for Docker Compose in development and Kubernetes in production. SDKs are available for Python and TypeScript to create and manage collections programmatically.
Use Cases
Airweave reduces the engineering effort needed to make organizational data queryable by agents by providing prebuilt connectors, an extraction-to-embedding pipeline, and hosted search endpoints. Teams can centralize content from multiple tools and databases into a single semantically searchable index, keep data fresh with incremental syncs, and support multiple tenants or branded offerings with white-labeling. The REST and MCP interfaces plus language SDKs make it straightforward to integrate search into agent workflows or applications. The included dashboard and Swagger API docs simplify configuration and testing, while the documented stack and deployment scripts enable local development and production deployments on Kubernetes for scale.

Please fill the required fields*