Jeremy
API Reference

Ingest URL

Ingest documentation directly from a URL.

POST /api/ingest-url

A simplified ingestion endpoint. Provide a URL and Jeremy will fetch the content, chunk it, and ingest it automatically. Supports llms.txt files and plain web pages.

If the library already exists, its chunks are replaced.

Auth: session (dashboard login)

Request Body

{
  "libraryId": "my-lib",
  "name": "My Library",
  "description": "Optional description",
  "sourceUrl": "https://example.com/llms.txt",
  "sourceType": "llms_txt"
}
FieldTypeRequiredDescription
libraryIdstringYesUnique identifier for the library
namestringYesDisplay name
descriptionstringNoLibrary description
sourceUrlstringYesURL to fetch documentation from
sourceTypestringYesContent type. Use llms_txt for llms.txt files, or another value for plain pages

Response

{
  "success": true,
  "libraryId": "my-lib",
  "chunksIngested": 87,
  "vectorized": true
}

Example

curl -X POST https://jeremy-app.ian-muench.workers.dev/api/ingest-url \
  -H "Content-Type: application/json" \
  -H "Cookie: your_session_cookie" \
  -d '{
    "libraryId": "react",
    "name": "React",
    "sourceUrl": "https://react.dev/llms.txt",
    "sourceType": "llms_txt"
  }'

Errors

StatusDescription
400Missing required fields, or URL could not be fetched
401Not authenticated (session required)