Start a crawl

Sends the crawler to a site. Every scraped page lands as an ordinary web document while the crawl runs. Scraping a site again updates the pages already in the library and asks for a live copy. Up to 3 crawls at once, and the same site cannot be crawled twice in parallel.

POST/knowledge/crawls

Authorization

AuthorizationBearer token · headerrequired

`MEDIAN_KEY` from Settings under API, or an MCP OAuth access token. A key acts as the organization; a token acts as the person who approved it.

Request body

requiredapplication/json

urlstringrequired

Responses

202Created.

idstring

400The request is malformed, and the message names the field.

errorobject

Show properties

codestring

messagestring

401The bearer token is missing, revoked, or expired.

errorobject

Show properties

codestring

messagestring

409It already exists, or it was already settled.

errorobject

Show properties

codestring

messagestring

429Too many requests. Wait the seconds in `Retry-After`. Limits depend on the plan. See [rate limits](/api/errors-and-limits#rate-limits).

errorobject

Show properties

codestring

messagestring

Request

curl -X POST "https://api.median.sh/v1/knowledge/crawls" \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
  "url": "docs.example.com"
}'

const response = await fetch("https://api.median.sh/v1/knowledge/crawls", {
  method: "POST",
  headers: {
    "Authorization": "Bearer YOUR_TOKEN",
    "Content-Type": "application/json"
  },
  body: JSON.stringify({
  "url": "docs.example.com"
})
});

Response

{
  "id": "string"
}