# Nanonets OCR — how to use (mcp.ai)

Connect your Nanonets OCR account and use 6 tools for document data extraction straight from your AI agent. Connect with your own API key. Nanonets OCR extracts structured data from documents using trained, model-based OCR workflows.

## Option A — via MCP (recommended)
Remote MCP endpoint (HTTP, streamable): `https://api.mcp.ai/p_nanonets_ocr?ms=1787295960000`
Add it as a custom/remote MCP connector in your client (Claude, Cursor, VS Code…), then authenticate when prompted. Once connected, ask the agent to use the server's tools (e.g. `nanonets_ocr_extract_from_url`).

## Option B — via direct REST API
Base URL: `https://api.mcp.ai/api/nanonets_ocr`
Auth: `Authorization: Bearer sk_live_…` — create a workspace API key at https://mcp.ai/settings/api-keys
Discover endpoints: `GET https://api.mcp.ai/api/nanonets_ocr/_endpoints`

### Endpoints
- `POST https://api.mcp.ai/api/nanonets_ocr/extract/from/url` — Synchronously extract model-defined fields and tables from one publicly reachable document URL. Returns durable page and file IDs even when the model finds no matching fields.
  - body: { url: string, model_id: string, request_metadata?: string }
- `POST https://api.mcp.ai/api/nanonets_ocr/get/model` — Get readiness, accuracy, extraction fields, and table configuration for one Nanonets OCR model.
  - body: { model_id: string }
- `POST https://api.mcp.ai/api/nanonets_ocr/get/prediction/file` — Get normalized extraction results for every page of one previously processed Nanonets OCR file. A successful page can have no extracted fields.
  - body: { model_id: string, request_file_id: string }
- `POST https://api.mcp.ai/api/nanonets_ocr/get/prediction/page` — Get normalized extracted fields and tables for one OCR prediction page.
  - body: { page_id: string, model_id: string }
- `POST https://api.mcp.ai/api/nanonets_ocr/list/models` — List the OCR models available to the connected Nanonets account, including each model's ID, readiness state, and extraction categories.
- `POST https://api.mcp.ai/api/nanonets_ocr/list/predictions` — List compact OCR prediction-page summaries for a model within an inclusive UTC date range. Nanonets returns the entire matching range without pagination, so prefer narrow ranges when a model has subst
  - body: { end_date: string, model_id: string, start_date: string }

## Example prompts
- "What can I do in Nanonets OCR?"
- "Show me a summary of my Nanonets OCR account"

## More
- Page: https://mcp.ai/nanonets_ocr
- Agent spec (llms.txt): https://mcp.ai/nanonets_ocr/llms.txt
- Postman collection: https://mcp.ai/nanonets_ocr/postman.json