Airweave is an open-source tool that lets agents search any app. It connects to productivity tools, databases, or document stores and transforms their contents into searchable knowledge bases for agents.
TL;DR: Airweave transforms app data into agent-ready knowledge.
š¬ Demo
Check out our demo below to see Airweaveās G-Drive and Slack search in action used by Claude Desktop, showcasing how Airweave outperforms Anthropicās own built-in Google Drive connector ⤵ļø
Agents struggle with vague or informal requestsĀ likeĀ āResolve that one Asana ticket about auth configsā or ācancel Alexās Stripe payment for Taylor Swift ticketsā
Typically, then agents hallucinate when they can't find data, or rely on inefficient chains of function calls
MCP encourages agent-app interactions, but MCP servers alone donāt make workspaces semantically searchable
ā Our Solution: Airweave
Airweave connects to productivity tools, databases, or document stores makes their contents searchable for agents
Bottom line:Ā Your agent gets a simple search endpoint. Your users get reliable, personalized experiences
Features
Open-Source
Simple yet powerful search interfaceĀ via REST or MCP
MCP-compatible: Turn any app into a semantically searchable MCP server
Connect to your agent to your favorite apps like Google Drive, Gmail, Outlook, Google Calendar, HubSpot, Notion, Jira, Slack, Dropbox, Stripe, Github, etc.
Legal AI assistants searching Google Drive/OneDrive to give accurate Perplexity-like answers
Agents for engineering managers scanning GitHub, Notion, and Jira to understand codebases and draft design docs
Compliance agents verifying marketing materials against current Dropbox financial data in regulated industries
š¤ The Team
Lennert and Rauf met in uni 7 years ago, and have been friends since. Lennert did AI research at Amazon and IBM, and has been working with LLMs since 2020. Rauf is a CS grad and worked as a Data Platform Engineer at fast-growing startups and large enterprise.