{"name":"Airweave: Context Retrieval for AI Agents Across Apps and Databases","description":"Airweave is an open-source context retrieval layer designed for AI agents, enabling them to access information across various applications and databases. It transforms diverse content into searchable knowledge bases, offering a standardized interface for agents to perform semantic, hybrid, and recency-biased searches. The platform simplifies data synchronization, entity extraction, and serves as a robust foundation for building intelligent AI applications.","github":"https://github.com/airweave-ai/airweave","url":"https://osrepos.com/repo/airweave-ai-airweave","source":"osrepos.com","sourceDescription":"This repository profile is provided by osrepos.com, an open source repository discovery platform.","repositoryProfile":"https://osrepos.com/repo/airweave-ai-airweave","generatedFor":"open source discovery and AI-assisted research","markdown":"https://osrepos.com/repo/airweave-ai-airweave.md","json":"https://osrepos.com/repo/airweave-ai-airweave.json","topics":["AI Agents","Context Retrieval","LLM","RAG","Vector Database","Knowledge Graph","Python","Data Synchronization"],"keywords":["AI Agents","Context Retrieval","LLM","RAG","Vector Database","Knowledge Graph","Python","Data Synchronization"],"stars":null,"summary":"Airweave is an open-source context retrieval layer designed for AI agents, enabling them to access information across various applications and databases. It transforms diverse content into searchable knowledge bases, offering a standardized interface for agents to perform semantic, hybrid, and recency-biased searches. The platform simplifies data synchronization, entity extraction, and serves as a robust foundation for building intelligent AI applications.","content":"## Introduction\n\nAirweave is a fully open-source context retrieval layer designed to empower AI agents by providing them with access to information across various applications and databases. It connects seamlessly to productivity tools, databases, and document stores, transforming their diverse contents into semantically searchable knowledge bases. This platform offers a standardized interface, accessible via REST API or MCP, handling everything from authentication and data extraction to embedding and serving, making it easier for agents to find relevant information.\n\n## Installation\n\nYou can get started with Airweave quickly by self-hosting it using Docker and Docker Compose.\n\nFirst, ensure Docker and Docker Compose are installed on your system.\n\nbash\n# 1. Clone the repository\ngit clone https://github.com/airweave-ai/airweave.git\ncd airweave\n\n# 2. Build and run\nchmod +x start.sh\n./start.sh\n\n\nOnce the setup is complete, you can access the Airweave dashboard at `http://localhost:8080`. For a managed service option, consider [Airweave Cloud](https://app.airweave.ai/).\n\n## Examples\n\nAirweave provides SDKs for both Python and TypeScript/JavaScript, allowing developers to easily integrate its powerful search capabilities into their applications.\n\n### Python SDK\n\nInstall the SDK:\n\nbash\npip install airweave-sdk\n\n\nExample usage:\n\npython\nfrom airweave import AirweaveSDK\n\n# Initialize client\nclient = AirweaveSDK(\n    api_key=\"YOUR_API_KEY\",\n    base_url=\"http://localhost:8001\"\n)\n\n# Create a collection\ncollection = client.collections.create(name=\"My Collection\")\n\n# Add a source connection (e.g., Stripe)\nsource = client.source_connections.create(\n    name=\"My Stripe Connection\",\n    short_name=\"stripe\",\n    readable_collection_id=collection.readable_id,\n    authentication={\n        \"credentials\": {\"api_key\": \"your_stripe_api_key\"}\n    }\n)\n\n# Semantic search (default)\nresults = client.collections.search(\n    readable_id=collection.readable_id,\n    query=\"Find recent failed payments\"\n)\n\n# Hybrid search (semantic + keyword)\nresults = client.collections.search(\n    readable_id=collection.readable_id,\n    query=\"customer invoices Q4 2024\",\n    search_type=\"hybrid\"\n)\n\n# Get AI-generated answer instead of raw results\nanswer = client.collections.search(\n    readable_id=collection.readable_id,\n    query=\"What are our customer refund policies?\",\n    response_type=\"completion\",\n    enable_reranking=True\n)\n\n\n### TypeScript/JavaScript SDK\n\nInstall the SDK:\n\nbash\nnpm install @airweave/sdk\n# or\nyarn add @airweave/sdk\n\n\nExample usage:\n\ntypescript\nimport { AirweaveSDKClient, AirweaveSDKEnvironment } from \"@airweave/sdk\";\n\n// Initialize client\nconst client = new AirweaveSDKClient({\n    apiKey: \"YOUR_API_KEY\",\n    environment: AirweaveSDKEnvironment.Local\n});\n\n// Create a collection\nconst collection = await client.collections.create({\n    name: \"My Collection\"\n});\n\n// Add a source connection (e.g., Stripe)\nconst source = await client.sourceConnections.create({\n    name: \"My Stripe Connection\",\n    shortName: \"stripe\",\n    readableCollectionId: collection.readableId,\n    authentication: {\n        credentials: { apiKey: \"your_stripe_api_key\" }\n    }\n});\n\n// Semantic search (default)\nconst results = await client.collections.search(\n    collection.readableId,\n    { query: \"Find recent failed payments\" }\n);\n\n// Hybrid search (semantic + keyword)\nconst hybridResults = await client.collections.search(\n    collection.readableId,\n    {\n        query: \"customer invoices Q4 2024\",\n        searchType: \"hybrid\"\n    }\n);\n\n// Get AI-generated answer instead of raw results\nconst answer = await client.collections.search(\n    collection.readableId,\n    {\n        query: \"What are our customer refund policies?\",\n        responseType: \"completion\",\n        enableReranking: true\n    }\n);\n\n\n## Why Use Airweave\n\nAirweave stands out as a powerful solution for enhancing AI agent capabilities through robust context retrieval. Its key features include:\n\n*   **Extensive Data Synchronization**: Connects to over 30 sources, including popular apps like GitHub, Notion, Slack, Salesforce, and various databases, with minimal configuration.\n*   **Intelligent Data Processing**: Features entity extraction and a sophisticated transformation pipeline to prepare data for optimal search.\n*   **Scalable Architecture**: Built with a multi-tenant architecture and OAuth2 support, ensuring security and scalability.\n*   **Efficient Updates**: Utilizes content hashing for incremental updates, keeping knowledge bases fresh without redundant processing.\n*   **Advanced Search Capabilities**: Offers semantic search for nuanced agent queries, alongside hybrid search (semantic + keyword), query expansion, reranking, and recency bias options.\n*   **Data Versioning**: Provides versioning for data changes, ensuring traceability and reliability.\n\nWith its comprehensive feature set and support for a wide array of integrations, Airweave simplifies the creation of intelligent AI applications that can effectively leverage vast amounts of distributed information.\n\n## Links\n\n*   **GitHub Repository**: [airweave-ai/airweave](https://github.com/airweave-ai/airweave)\n*   **Official Documentation**: [Airweave Docs](https://docs.airweave.ai/welcome)\n*   **Discord Community**: [Join Discord](https://discord.com/invite/484HY9Ehxt)\n*   **Twitter**: [Follow @airweave_ai](https://x.com/airweave_ai)","metrics":{"detailViews":4,"githubClicks":3},"dates":{"published":null,"modified":"2025-12-19T12:01:18.000Z"}}