> ## Documentation Index
> Fetch the complete documentation index at: https://docs.minns.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Search claims

> Semantic search over extracted claims — find soft facts like user preferences and intent.

Use this endpoint for fuzzy, concept-based queries like *"What are the user's favorite genres?"* or *"Has the user complained before?"*

## Request body

<ParamField body="query_text" type="string" required>
  Natural language query. Keep it conversational — don't include IDs.
</ParamField>

<ParamField body="top_k" type="integer" default="10">
  Number of results to return.
</ParamField>

<ParamField body="min_similarity" type="number" default="0.7">
  Minimum similarity threshold (`0.0`–`1.0`).
</ParamField>

### Example request

```json theme={null}
{
  "query_text": "What are the user's favorite genres?",
  "top_k": 3,
  "min_similarity": 0.7
}
```

## Response

Returns an array of `ClaimResponse` objects with the `similarity` field populated.

```json theme={null}
[
  {
    "claim_id": 42,
    "claim_text": "User prefers Sci-Fi movies",
    "confidence": 0.92,
    "source_event_id": 123456789012345678901234567890,
    "similarity": 0.89,
    "evidence_spans": [
      {
        "start_offset": 15,
        "end_offset": 45,
        "text_snippet": "I really love Sci-Fi movies"
      }
    ],
    "support_count": 3,
    "status": "Active",
    "created_at": 1738425600000000000,
    "last_accessed": 1738512000000000000
  }
]
```

<Tip>
  Keep the `query_text` natural. Don't include IDs in the string — filter the results in your application code instead.
</Tip>