semantic_search
Search a table's embedding column by query string using dense ANN indexing to return top matching rows ranked by vector similarity.
Instructions
Dense ANN semantic search using a query string (requires embedding column + ANN index). Prefers MongrelDB 0.64 native retrieve_text with semantic-identity provenance; falls back to SQL ann_search_exact.
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"table": "<table>",
"query": "<query>"
},
"name": "semantic_search"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| k | No | ||
| query | Yes | ||
| table | Yes | ||
| min_score | No | Cosine similarity floor; drops weak hits (exact path only) | |
| projection | No | ||
| provider_id | No | Required when the embedding column is application-supplied; must match the vector model | |
| exact_rerank | No | ||
| embedding_column | No | embedding |