AnchorGrid Intelligence API

Turn drawings and specs into data you can build on.

Upload a construction PDF, choose what you want to find, and get structured JSON back. Detect doors and elevations, pull schedules from drawing sets, and extract specification sections, then use the results in your own software.

Try it in the Playground first: no code or API key needed. Or browse the API Reference.

drawings.pdf · floor planPDF
Detect doors
Job completeJSON
{
  "status": "complete",
  "model": "door-detector",
  "result": { "doors": [ … ] }
}

Hello, World

Your first drawing. Your first detection.

Run a quick analysis in the Playground to see how AnchorGrid performs on your document—no code or API key required.

Make your first request

Start in the Playground—a simple, point-and-click way to upload a document, run an analysis, and inspect the results before building an integration.

  • A drawing or specification you want to test
  • No code or API key required
  1. Step 1

    Upload your document

    Upload a supported drawing or specification in the Playground.

  2. Step 2

    Choose your analysis

    Select the available detection or extraction capability you want to test.

  3. Step 3

    Run your analysis

    Run the test and inspect the detections and available structured results.

Compare the results with what you'd mark on the drawing by hand. That's the quickest way to judge whether the model fits your documents.

Once you've tested the model, connect it to your workflow. In production, your software or AI agent sends requests to AnchorGrid and receives structured JSON results automatically. Use the Playground to explore the model, then use the API to integrate it.

Choose your path

Prepare to use the AnchorGrid API in your software or with your AI agent.

Use the API in your software

Your application or backend sends documents and analysis requests to AnchorGrid, then reads the structured JSON results and uses them in your product or workflow.

  • Set up your API key, upload documents, start analyses and retrieve results with the REST API
  • Examples in cURL, Python and JavaScript

Requires an API key

Use the API with your AI agent

Your developers give the agent a tool that calls the AnchorGrid REST API. The agent decides when to analyze a document, and the tool returns the structured JSON for the agent to work with.

  • Uses the same REST API and X-API-Key authentication as your software
  • The API key stays in your integration's secure configuration, never in the chat

Requires an API key

Production runs on its own. Once connected, your software or agent sends each document and collects the results automatically. No one needs to upload files or click through the Playground for each request.

Searches these docs: guides, the basics and every endpoint.

Make your first requestPlayground walkthrough: upload, choose an analysis, run it
Connect your softwareYour application calls the REST API. Requires an API key
Connect your AI agentGive an agent a tool that calls the REST API
AuthenticationX-API-Key header, 401 errors, keeping keys safe
Detect doorsPOST /v1/drawings/detection/doorsDoor detection — Find the doors on the pages of an architectural drawing set.
Check your resultsGET /v1/jobs/{job_id}Get a job — See a job’s status and, once it’s complete, its results.