Docu Scan MCP
Extract text and structure from PDFs and documents in your agent pipeline using Google Document AI without building OCR infrastructure yourself.
Overview
Docu-Scan MCP is an MCP server for the Build phase that extracts PDF and document content via Google Document AI behind a Bearer-authenticated remote.
What is this MCP server?
- PDF and document extraction powered by Google Document AI
- Streamable-http MCP at docu-scan.spocont.com with required Bearer authorization
- Free trial available via POST /api/document-scan/access/purchase for access tokens
- Catalog version 1.0.0 focused on scan-and-extract workflows
- Keeps heavy OCR out of your repo—agent calls the hosted Docu-Scan remote
- MCP version 1.0.0 on streamable-http
- Required Authorization Bearer header per catalog schema
- Free trial available via documented access purchase endpoint
What problem does it solve?
Shipping document-heavy features stalls when every PDF needs manual parsing or a bespoke Google Cloud Document AI integration.
Who is it for?
Builders who want Document AI–grade extraction in the IDE with a hosted MCP and a trial token path.
Skip if: Teams that cannot send documents to a third-party remote or need fully on-prem OCR with no HTTP egress.
What do I get? / Deliverables
Your agent returns extracted document text and fields through MCP calls so you can build RAG, forms, and workflows on real files faster.
- Extracted document content usable in agent-driven codegen and tests
- Faster RAG or form-mapping prototypes on real PDFs
- Document AI-backed parsing without self-hosting processors
Recommended MCP Servers
Journey fit
Document extraction is a Build-time integration when ingest pipelines, RAG, or back-office automations need reliable parsed files. Docu-Scan is an external Document AI gateway exposed as MCP tools, not a ship-phase security audit product.
How it compares
Google Document AI extraction MCP with token billing, not a local PDF markdown skill.
Common Questions / FAQ
Who is Docu Scan MCP for?
Docu-Scan is for solo builders and agents automating PDF and document ingest who prefer a hosted Document AI MCP over rolling their own GCP pipeline.
When should I use Docu Scan MCP?
Use it during Build integrations when prototypes need real extractions—contracts, invoices, onboarding packets—before you commit to a long-term OCR architecture.
How do I add Docu Scan MCP to my agent?
Obtain a Bearer token from POST /api/document-scan/access/purchase on docu-scan.spocont.com, add https://docu-scan.spocont.com/api/document-scan/mcp as a streamable-http remote, and set the Authorization header as a secret in your MCP client.