Transform MCP
VerifiedFeaturedRemoteAbout Transform MCP
Transform MCP is a remote document-processing server for agents, turning files into structured, enriched, chunked, and embedded data for RAG, vector stores, or agent memory. It handles more than 60 file formats through a single call.
Key features
- 01Processes over 60 file formats
- 02Partitions and enriches document content
- 03Chunks and embeds output
- 04Supports extraction jobs and schemas
- 05Checks job status and retrieves results
Use cases
- 01Preparing PDFs and scans for RAG
- 02Structuring spreadsheets or email for agent memory
- 03Extracting data from files using a suggested schema
MCP support
It exposes a remote, streamable-HTTP MCP server at https://mcp.transform.unstructured.io. Tools support file upload requests, transformation and extraction jobs, schema suggestions, status checks, and result retrieval.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://mcp.transform.unstructured.ioclaude mcp add --transport http 23c6ec https://mcp.transform.unstructured.io{"mcpServers":{"23c6ec":{"type":"http","url":"https://mcp.transform.unstructured.io"}}}Open in Cursor{"servers":{"23c6ec":{"type":"http","url":"https://mcp.transform.unstructured.io"}}}Open in VS Code{"mcpServers":{"23c6ec":{"type":"http","url":"https://mcp.transform.unstructured.io"}}}{"mcpServers":{"23c6ec":{"serverUrl":"https://mcp.transform.unstructured.io"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
Details
From the project READMEunstructured-mcp-integrations
Integration templates and runnable examples for the Unstructured Transform MCP server across registries, marketplaces, and agent frameworks. This is the canonical, public home for these integrations.
What Transform MCP is
Transform MCP turns any file into agent-ready data, called directly from your agent session with no separate pipeline to wire up. Drop in a PDF, spreadsheet, scan, or email and get back partitioned, enriched, chunked, and embedded output ready for RAG, vector stores, or agent memory. 60+ formats, one call.
Connection reference
| Field | Value |
|---|---|
| Endpoint | https://mcp.transform.unstructured.io |
| Transport | Remote, streamable-HTTP (connect directly, or via npx mcp-remote for clients without native remote-MCP support) |
| Tools | request_file_upload_url, start_transform_job, start_extraction_job, suggest_extraction_schema_for_file, check_job_status, get_job_results |
| Docs | https://docs.unstructured.io/transform/overview |
| Support | [email protected] |
| Security | https://trust.unstructured.io |
Integrations
One folder per integration under transform/. Copy
_template/ to add one. This table is the index of what exists and
where each one stands.
| Integration | Category | Auth | Status | Owner |
|---|---|---|---|---|
| Cline | marketplace | OAuth / API key | Not started | @simoncoombes |
| Vercel AI SDK | agent-framework | API key | Not started | @mikedownesdev |
| Cursor | marketplace | OAuth / API key | Not started | @ctrahey |
| Grok Build | marketplace | OAuth | Not started | @nisha-pm |
Adding an integration
See CONTRIBUTING.md. In short: copy _template/ into
transform/<name>/, fill in the README, example, assets, and meta.yaml, then
open a PR.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.