Search Documentation
Search across all documentation pages
semantic_search
Search notes by meaning using vector embeddings. Finds conceptually similar content even without exact keyword matches. Requires OpenAI API key in config. For keyword-based search, use search_notes instead.
Parameters#
| Parameter | Type | Required | Description |
|---|---|---|---|
| query | string | required | Natural language search query |
| topK | integer (1-50) | optional | Max results (default 10) |
| threshold | number (0-1) | optional | Minimum similarity threshold 0-1 (default 0.5) |
| vaultId | string (UUID) | optional | Vault to search within. Uses default vault if omitted. |
Returns#
An array of note objects ranked by semantic similarity, each containing the note ID, title, content preview, similarity score, and vault info.
Example#
json
{
"tool": "semantic_search",
"arguments": {
"query": "how to handle authentication tokens securely",
"topK": 5,
"threshold": 0.6
}
}