Quickstart

1. Get an API key

Sign in at the Scout dashboard and create a key under Settings → API keys. Keep it server-side — never embed it in browser, mobile, or other untrusted code.

2. Send a request

Every endpoint accepts a JSON body over HTTPS. Pass your key as a Bearer token in the Authorization header.

$
curl -X POST https://core.usescout.sh/v1/search \
>
  -H "Authorization: Bearer YOUR_API_KEY" \
>
  -H "Content-Type: application/json" \
>
  -d '{"queries": ["open source vector databases"], "limit": 5}'

3. Read the response

Responses are flat JSON. A search returns a list of results, each with the citation URL, a title, an excerpt, and an optional publish_date. Top-level fields include a search_id, a session_id (echoed or generated) and a usage breakdown by SKU. The credits field reports the relative cost of the call.

1
{
2
  "search_id": "search_4f5e...",
3
  "session_id": "sess_b8c1...",
4
  "queries": ["open source vector databases"],
5
  "results": [
6
    {
7
      "url": "https://example.com/vector-db-guide",
8
      "title": "A Guide to Open Source Vector Databases",
9
      "excerpts": [
10
        "Vector databases index high-dimensional embeddings for fast similarity search."
11
      ],
12
      "publishedDate": "2024-12-02",
13
      "publish_date": "2024-12-02"
14
    }
15
  ],
16
  "warnings": null,
17
  "credits": 1,
18
  "usage": [{ "name": "sku_search", "count": 1 }]
19
}

Want more in one call?

Pass advanced_settings.contents to fetch each result’s full text, focused highlights and a per-result summary in a single request:

1
{
2
  "queries": ["AI funding rounds 2026"],
3
  "objective": "latest Series A/B AI funding",
4
  "category": "news",
5
  "advanced_settings": {
6
    "source_policy": { "after_date": "2026-01-01" },
7
    "contents": {
8
      "text": true,
9
      "highlights": { "query": "funding amount, investors" },
10
      "summary":    { "query": "Who raised, how much, from whom?" }
11
    }
12
  }
13
}

4. Pick the right endpoint