airweave
by lennertjansen
Productivity & TasksBrief
Context retrieval layer for AI agents across users' applications.
Description
What This Skill Does
Airweave is a context retrieval layer that indexes and syncs data from connected apps so AI agents can search it. Supports semantic, keyword, and hybrid retrieval strategies. Agents query collections to answer questions grounded in the user's actual workspace data.
Instead of switching between apps to find context, agents retrieve and synthesize relevant data from all connected sources in a single query.
When to Use It
- Finding which Slack thread discussed a specific product decision
- Locating a Confluence policy doc without knowing its exact title
- Checking Jira for open bugs related to a feature before shipping
- Pulling recent Salesforce notes on a customer account
- Answering who owns a project based on Linear tickets
Example Workflow
User asks: What did we decide about the refund policy in Notion?
- 1
Identifies query as a workspace data lookup requiring search
- 2
Formulates query: 'refund policy decision'
- 3
Runs: python3 scripts/search.py 'refund policy decision' --strategy hybrid
- 4
Evaluates result scores to determine confidence level
- 5
Synthesizes relevant passages and cites the source document
A direct answer referencing the Notion document where the refund policy was documented, with source attribution.
Requirements
- AIRWEAVE_API_KEY environment variable
- AIRWEAVE_COLLECTION_ID environment variable
- Airweave account with at least one configured collection
Installation
clawhub install lennertjansen/airweaveRun the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.