API

ocr

Runs in your own browser. Recognises text in photos, scans and PDFs and exports it as plain text, Markdown or JSON with layout data. The most useful call in the catalogue for a model that has to work with paper: everything else downstream can read text, but nothing can read a scan.

Call it

POST https://timbrica.com/v1/tools/ocr/runs
Authorization: Bearer $TIMBRICA_KEY
Idempotency-Key: <your retry-safe key>
Content-Type: application/json

Input

Browser recipe

This tool runs in your own browser. The work order you receive carries this recipe with the placeholders filled in; the selectors below are a versioned contract, not implementation detail — they do not change without a version bump.

{
    "engine": "dom",
    "steps": [
        {
            "action": "upload",
            "selector": "[data-tmb-api=\"input-file\"]",
            "accept": [
                ".jpg",
                ".jpeg",
                ".png",
                ".webp",
                ".tif",
                ".tiff",
                ".bmp",
                ".pdf"
            ]
        },
        {
            "action": "click",
            "selector": "[data-tmb-api=\"run\"]"
        },
        {
            "action": "wait",
            "selector": "[data-tmb-api=\"download\"][data-tmb-format=\"{{format}}\"]",
            "timeout_s": 600,
            "appears": "after_run"
        },
        {
            "action": "click",
            "selector": "[data-tmb-api=\"download\"][data-tmb-format=\"{{format}}\"]",
            "appears": "after_run"
        }
    ],
    "result": {
        "kind": "download",
        "settles": "automatically"
    }
}

Errors it can return

validation_failed  insufficient_tokens 

History

• v1 — 2026-08-02 — Initial release (browser lane).

This contract cannot change without a new version and a line above: a test compares the published schema, pricing and limits against a committed lockfile and fails the build otherwise. Subscribe to /v1/changelog to be told when something moves.