recall
Your primary search tool. Use it to query your vault for relevant knowledge across all entity types.
Parameters
| Parameter | Type | Required | Description |
|---|
query | string | Yes | Natural language search query |
depth | string | No | Search depth: shallow, standard, or deep. Default: standard |
entity_type | string | No | Filter by type: person, project, concept, technology, etc. |
limit | number | No | Maximum results to return. Default: 10 |
expand_graph | boolean | No | Include related entities in results. Default: false |
excerpt_limit | number | No | Max length of content excerpts. Default: 500 |
scope | string | No | Search scope: all, recent, starred |
Examples
Simple search:
“Recall anything about the payments migration”
The AI calls: recall(query: "payments migration")
Filtered search:
“Who have I been working with on the API project?”
The AI calls: recall(query: "API project collaborators", entity_type: "person")
Deep search with connections:
“Give me everything you know about our Kubernetes setup”
The AI calls: recall(query: "Kubernetes setup", depth: "deep", expand_graph: true)
Tips for good queries
- Be specific: “Stripe webhook error handling” works better than “Stripe”
- Use names: “Sarah’s feedback on the dashboard” finds more relevant results
- Include timeframes: “database decisions from last month” narrows results
- Combine concepts: “authentication + mobile app” finds intersection of topics
The expand_graph parameter is powerful for understanding context. It returns not just matching entities but their direct relationships, giving your AI a fuller picture.
