# Airparser — how to use (mcp.ai)

Connect your Airparser account and use 10 tools for document data extraction straight from your AI agent. Connect with your own API key. Airparser extracts structured data from emails, documents, and other files using AI-powered parsing.

## Option A — via MCP (recommended)
Remote MCP endpoint (HTTP, streamable): `https://api.mcp.ai/p_airparser?ms=1787291460000`
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. `airparser_clone_extraction_schema`).

## Option B — via direct REST API
Base URL: `https://api.mcp.ai/api/airparser`
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/airparser/_endpoints`

### Endpoints
- `POST https://api.mcp.ai/api/airparser/clone/extraction/schema` — Replace a destination inbox's extraction schema with a copy of a source inbox's schema.
  - body: { source_inbox_id: string, destination_inbox_id: string }
- `POST https://api.mcp.ai/api/airparser/create/inbox` — Create an Airparser inbox/parser for a document family using a text or vision extraction engine.
  - body: { name: string, llm_engine?: string }
- `POST https://api.mcp.ai/api/airparser/delete/inbox` — Permanently delete an Airparser inbox and every document it contains. This destructive, cascading operation cannot be undone and requires explicit confirmation.
  - body: { inbox_id: string, confirm_permanent_deletion: boolean }
- `POST https://api.mcp.ai/api/airparser/get/account/usage` — Return the connected Airparser account's role, verification state, plan, and parsing-credit usage without exposing personal, secret, or internal account fields.
- `POST https://api.mcp.ai/api/airparser/get/document` — Get parsed output and status for a document, optionally including extended source and navigation details.
  - body: { source?: string, document_id: string, detail_level?: string }
- `POST https://api.mcp.ai/api/airparser/get/inbox` — Get one Airparser inbox, including its current extraction schema and useful parsing configuration.
  - body: { inbox_id: string }
- `POST https://api.mcp.ai/api/airparser/list/documents` — List and search documents in one inbox, one page at a time.
  - body: { query?: string, to_date?: string, inbox_id: string, per_page?: integer, from_date?: string, next_cursor?: string }
- `POST https://api.mcp.ai/api/airparser/list/inboxes` — List the Airparser inboxes available to the connected account.
- `POST https://api.mcp.ai/api/airparser/parse/document/sync` — Upload one document to an inbox and wait for its extracted JSON. This non-idempotent operation consumes plan credits, including one credit per PDF page. Do not retry automatically after an ambiguous f
  - body: { file: object, inbox_id: string, metadata?: object }
- `POST https://api.mcp.ai/api/airparser/upsert/extraction/schema` — Create or completely replace the extraction schema used by an Airparser inbox. The fields array is the complete replacement, not a partial update.
  - body: { fields: object[], inbox_id: string }

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

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