List Queries | Captain

import requests
BASE_URL = "https://api.captain.dev"
API_KEY = "your_api_key"
headers = {
    "Authorization": f"Bearer {API_KEY}",
}
response = requests.get(
    f"{BASE_URL}/v2/queries",
    headers=headers,
    params={
        "limit": 25,
        "sort": "desc"
    },
    timeout=30.0
)
if response.status_code == 200:
    data = response.json()
    for q in data["queries"]:
        print(f"{q['created_at']}  {q['latency_ms']} ms  "
              f"{q['result_count']} results  {q['query_text']!r}")
    # Page until next_cursor is null
    print(f"next_cursor={data['next_cursor']}")
else:
    print(f"Error: {response.status_code}")

{
  "queries": [
    {
      "query_id": "019a0e2a-7b31-7c4e-9d2f-3a8c1e5b6f90",
      "query_text": "What are the adverse reactions listed for metformin?",
      "collection_name": "clinical-docs",
      "environment": "production",
      "status": "completed",
      "api_version": "v3",
      "created_at": "2026-09-05T18:46:25.660882Z",
      "latency_ms": 412,
      "result_count": 5,
      "request_id": "req_1788633985_84509ca6",
      "params": {
        "limit": 5,
        "semantic_ratio": 0.5,
        "has_filter": false,
        "has_boost": false,
        "exclude_chunk_types": null,
        "include": null,
        "rerank": true
      },
      "results": null,
      "request": null,
      "response": null,
      "body_status": null
    }
  ],
  "count": 1,
  "next_cursor": "eyJ2IjoxLCJrIjp7InF1ZXJ5X2lkIjoiMDE5YTBlMmEtN2IzMS03YzRlLTlkMmYtM2E4YzFlNWI2ZjkwIiwib3JnYW5pemF0aW9uX2lkIjoib3JnIiwiY3JlYXRlZF9hdCI6IjIwMjYtMDktMDVUMTg6NDY6MjUifSwicyI6ImRlc2MifQ"
}

List your stored queries, newest first, scoped to the environment of the API key you call with (a production key lists production queries).

Page with cursor until next_cursor is null; there is no total count. A page can be shorter than limit while next_cursor is still set when a filter trimmed it. Your key’s environment is part of the index, so pages come back full in either direction; collection and status are applied while reading, so a page can be shorter than limit while next_cursor is still set, and the API keeps reading to fill it. from and to bound the window on the index, so pair them with the other filters on large histories.

include inlines the same bodies GET /v2/queries/{query_id} returns, per row, so a feed and its previews can come from one request: results (retrieved chunks in one normalised shape), request, response. Each costs a stored-body read, so limit is capped at 25 when include is set. The four body fields are always present on every row and null unless requested.

History contains completed, non-streamed queries. Bodies are archived 90 days after the query (body_status: archived); the summary rows are kept.

Query parameters

collectionstring or nullOptional

Only queries against this collection.

statusstring or nullOptional

Filter by query status (completed or failed). Only completed queries are recorded today.

fromstring or nullOptional

Only queries created at or after this ISO-8601 date or timestamp (UTC when no offset is given).

tostring or nullOptional

Only queries created before this ISO-8601 date or timestamp (exclusive).

sortstring or nullOptionalDefaults to desc

Order by creation time: desc (newest first, default) or asc.

limitintegerOptional1-100Defaults to 25

Queries per page (1–100; at most 25 when include is set).

cursorstring or nullOptional

Continue from a previous page’s next_cursor.

includestring or nullOptional

Comma-separated bodies to inline on each row: results (normalised retrieved chunks), request, response. Each costs a stored-body read; without it a row is a summary.

Response

Successful Response

querieslist of objects

countinteger

next_cursorstring or nullOptional

Errors

400

Bad Request Error

401

Unauthorized Error

403

Forbidden Error

Headers

Query parameters

Response

Errors