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

# Research matches

> GET /api/matches: retrieve ranked cross-source research leads for a case.

```
GET /api/matches
```

**caseId** `string` *required*

The case to find candidates for, for example `namus-123`.

```bash theme={null}
curl "https://api.casedock.net/api/matches?caseId=namus-123" \
  -H "Authorization: Bearer YOUR_DEVELOPER_API_KEY"
```

The response includes ranked candidates and a note describing the comparison.

## Candidate fields

**score** `number`

Overall score from 0 to 100.

**confidence** `string`

`strong`, `moderate`, or `weak`. Limited by evidence strength and lowered by discrepancies, so demographic overlap alone is never strong.

**evidenceScore** `object`

`{ score, strength }` for overlapping physical and forensic details (marks, belongings, identifiers). `strength` is `high`, `moderate`, `low`, or `none`.

**breakdown** `array`

Points per factor. Each entry has `key`, `label`, `kind` (`cross-reference` or `evidence`), `points`, `max`, and `state` (`match`, `partial`, `none`, or `unknown` when a record lacks the data).

**coverage** `number`

Percentage of factors both records had data for.

**discrepancies** `array`

Details that point away from a match, such as a large age or height gap. They don't change `score`.

**defaultScore, tuned** `number, boolean`

Present only for signed-in members who set custom weights under **Tune scoring**. API-key requests always use the standard weights.

> [!WARNING]
>
> Ranked suggestions are research leads, not identifications. Verify any lead with the original agency or source.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.