Read a PDF from a URL.

curl https://getjson.de/https://www.w3.org/WAI/ER/tests/xhtml/testfiles/resources/pdf/dummy.pdf

{
  "source": {
    "url": "https://www.w3.org/WAI/ER/tests/xhtml/testfiles/resources/pdf/dummy.pdf",
    "finalUrl": "https://www.w3.org/WAI/ER/tests/xhtml/testfiles/resources/pdf/dummy.pdf",
    "kind": "file",
    "platform": "file",
    "provider": "file",
    "fetchedAt": "2026-10-27T21:24:00.392Z",
    "cached": false
  },
  "content": {
    "title": "dummy.pdf",
    "author": null,
    "publishedDate": null,
    "language": null,
    "engagement": {},
    "hashtags": [],
    "mentions": [],
    "linkedContent": [],
    "truncated": false,
    "text": "Dummy PDF file",
    "wordCount": 3
  },
  "meta": null,
  "wall": null,
  "result": null
}

Recorded from https://getjson.de on 2026-10-27 with no key. Long text is cut and ends in " …".

• The provider is file: getJSON downloads the file and reads its text layer.
• Without a key, or with a trial key that is not claimed yet, a file may have 5 MB; a larger one comes back with the wall too_large. With a claimed key or an account the limit is 50 MB.
• A scan or an image has no text layer and comes back with the wall no_text. Add a schema to read it with a model: without a key 3 times a day per network on this GET; with a key, send the link and the schema to POST /v1/extract (1 credit in the budget mode). model=ocr needs a claimed key or an account, because a trial key runs the budget mode only.
• A file on your disk: curl -F "file=@report.pdf" https://getjson.de/

curl -G https://getjson.de/https://www.w3.org/WAI/ER/tests/xhtml/testfiles/resources/pdf/dummy.pdf --data-urlencode 'schema={"type":"object","properties":{"title":{"type":"string"}}}'

Without a key a network has 3 schema calls a day. The call waits up to 60 s and fills result, validated against the schema. No answer of this call is recorded on this page.

The MCP tool read needs no key. Connect, then call it:

claude mcp add --transport http getjson https://getjson.de/mcp

{
  "name": "read",
  "arguments": {
    "url": "https://www.w3.org/WAI/ER/tests/xhtml/testfiles/resources/pdf/dummy.pdf"
  }
}

The tool answers with the same envelope as text.

getJSON reads any content a user or an agent points at, a web page or a file, and returns JSON in your schema, validated. Try and use it without an account. Guide for agents