0xPdf MCP
The 0xPdf MCP server enables AI agents to parse PDFs into structured JSON, manage schemas, handle async/batch jobs, and monitor usage via the 0xPdf API.
PDF Parsing
Synchronously parse a single PDF into JSON using a schema (by ID, name, JSON definition, or template), with optional OCR support
Parse a PDF via SSE streaming for real-time results
Submit a single PDF for asynchronous background processing (with optional webhook notification), poll until completion, or check job status at any time
Batch Processing
Submit multiple PDFs at once for async batch parsing
Check batch status/results and poll until all items finish
Schema Management
List all saved JSON schemas in your account
Create and save hand-written or AI-generated schemas
Auto-generate a JSON extraction schema from a natural language description using AI (costs wallet credits)
Refine existing schemas based on new descriptions or selected text
Account & Diagnostics
Check current billing status, wallet balance, and pricing
Verify backend connectivity via health check
Retrieve and reset MCP transport/runtime metrics for debugging and optimization
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@0xPdf MCP parse the PDF at ~/Downloads/invoice.pdf"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
0xPdf MCP
Schema in → PDF in → JSON out for Claude, Cursor, and other MCP clients.
This stdio MCP server wraps the 0xPdf HTTP API (auth, billing, rate limits intact). Agents can parse invoices/forms into a JSON schema, generate schemas, and poll async jobs.

Playground · Docs / MCP setup · Pricing
Install
pip install oxpdf-mcp
# or
uvx oxpdf-mcpGet an API key at 0xpdf.io/dashboard.
Related MCP server: PDF Figures MCP Server
Cursor / Claude Desktop
{
"mcpServers": {
"0xpdf": {
"command": "uvx",
"args": ["oxpdf-mcp"],
"env": {
"PDF_PARSING_API_BASE_URL": "https://api.0xpdf.io",
"PDF_PARSING_API_PREFIX": "api/v1",
"PDF_PARSING_API_KEY": "YOUR_API_KEY",
"PDF_PARSING_ALLOWED_API_HOSTS": "api.0xpdf.io",
"PDF_PARSING_REQUIRE_HTTPS": "true",
"PDF_PARSING_ALLOWED_FILE_ROOTS": "/home/YOU/Downloads:/home/YOU/Documents",
"PDF_PARSING_MCP_DISALLOW_FULL_RESPONSE_MODE": "true"
}
}
}
}Or with pip:
"command": "oxpdf-mcp"Tools (selected)
Tool | Purpose |
| Schema-first sync parse |
| Streaming parse (SSE) |
| Async jobs |
| Saved schemas |
| AI schema generation (wallet debit) |
| Balance / pricing |
| Liveness |
Full list and env vars: see source oxpdf_mcp/server.py and docs.
Env
Variable | Default | Notes |
| — | Required for authenticated tools |
|
| Production API |
| — | Required in production ( |
|
| Set |
| CWD | Colon-separated absolute paths for |
SDKs (non-MCP)
Python:
oxpdfJS/TS:
@0xpdf/client
License
MIT
Available Tools
15 toolscreate_schemaA
Save a JSON schema to your account (POST /api/v1/schemas). Requires API key. No wallet charge per save; subject to max saved schemas limit. Use after generate_schema_from_description or for hand-written schemas.
Inputs: name (required), schema_def (required), is_default, response_mode, +1 more. Returns: object
generate_schema_from_descriptionA
AI-generate a JSON schema from a description (POST /api/v1/schemas/generate). Requires API key, payment method, and prepaid wallet balance. Debits the same wallet ledger as PDF/OCR parsing on each successful generation. Pair with create_schema to persist.
Inputs: description (required), refinement, response_mode, selected_text, +2 more. Returns: object
get_mcp_runtime_metricsA
Get MCP runtime transport metrics for optimization/debugging.
Inputs: no named parameters. Returns: object
get_parse_batch_statusC
Get status for an async parse batch created by submit_parse_batch.
Inputs: batch_id (required), item_limit, include_items, response_mode, +2 more. Returns: object
get_parse_job_statusC
Get status for an async parse job created by submit_parse_job.
Inputs: job_id (required), response_mode, include_result, max_estimated_tokens_override. Returns: object
get_pricing_currentC
Get current billing/pricing status for the authenticated user.
Inputs: response_mode, max_estimated_tokens_override. Returns: object
health_checkA
Health check for backend connectivity. Useful before running parse tools.
Inputs: no named parameters. Returns: object
list_schemasC
List user schemas from /api/v1/schemas.
Inputs: limit, offset, response_mode, max_estimated_tokens_override. Returns: object
parse_pdf_streamC
Parse one PDF as SSE stream through /api/v1/pdf/parse-stream.
Inputs: pdf_path (required), pages, use_ocr, schema_id, +9 more. Returns: object
parse_pdf_syncC
Parse one PDF synchronously through /api/v1/pdf/parse.
Inputs: pdf_path (required), pages, use_ocr, schema_id, +6 more. Returns: object
reset_mcp_runtime_metricsA
Reset MCP runtime transport metrics counters.
Inputs: no named parameters. Returns: object
submit_parse_batchC
Submit multiple PDFs for async batch parsing via /api/v1/pdf/batch.
Inputs: pdf_paths (required), use_ocr, schema_id, ocr_engine, +4 more. Returns: object
submit_parse_jobC
Submit one PDF for async processing via /api/v1/pdf/upload.
Inputs: pdf_path (required), use_ocr, schema_id, ocr_engine, +4 more. Returns: object
wait_for_batch_completionA
Poll batch status with exponential backoff and jitter until it is completed/failed/partial or times out.
Inputs: batch_id (required), jitter_factor, timeout_seconds, max_interval_seconds, +1 more. Returns: object
wait_for_job_completionA
Poll job status with exponential backoff and jitter until it is completed/failed or times out.
Inputs: job_id (required), jitter_factor, timeout_seconds, max_interval_seconds, +1 more. Returns: object
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 15 tools
Each tool has a clearly distinct purpose. PDF parsing tools differentiate by sync/stream/async modes, schema tools are separate from parsing, and status, metrics, health, and pricing tools are all unique with no overlap.
All tools follow a consistent verb_noun pattern in lowercase snake_case, e.g., create_schema, get_parse_job_status, wait_for_batch_completion. No mixed conventions or vague verbs.
15 tools is well-scoped for the domain of PDF parsing and schema management. Each tool serves a necessary role without bloat.
Covers core workflows: PDF parsing (sync, stream, async), schema generation/persistence, status polling, and administrative tools. Minor gap: no schema update or delete, and no cancel/retry for async jobs.
Maintenance
Related MCP Connectors
Turn any PDF into structured JSON via AI + OCR: invoices, bank statements, contracts.
Turn PDFs and photos into typed JSON with your own schema. Missing values come back null.
Agent-native document parsing: PDF, scans and FR/EU invoices to structured JSON or Markdown.
Pay-per-call document and web tools for AI agents: PDF parsing and OCR (Japanese-strong), web and structured extraction, invoice verification, company profiles and PDF utilities. x402 (USDC on Base), no API key; failed calls are never charged.
Related MCP Servers
- AlicenseAqualityBmaintenanceEnables AI assistants to parse and convert PDFs and images to structured text formats using the Doc2x v2 API. Supports asynchronous document processing, format conversion, and file downloads with configurable polling and timeout settings.1465 npm3MIT
- AlicenseNot gradedqualityDmaintenanceExtracts figures and tables from PDF documents via a FastAPI service, wrapping PDFFigures 2.0. Enables AI agents to programmatically retrieve structured figure and table data from scholarly PDFs.9Apache 2.0
- AlicenseNot gradedqualityCmaintenanceExtracts text and tables from PDFs for AI agents via MCP, enabling structured data retrieval from invoices, reports, and statements.49 PyPI1MIT
- FlicenseNot gradedqualityDmaintenanceEnables AI agents to parse and analyze PDF invoices, including text, field, and table extraction, OCR for scanned documents, ZUGFeRD support, compliance validation, and batch processing via MCP.-