Extend MCP Server
Connect to the Extend MCP server to extract structured data, classify and split documents, parse files for AI, and monitor runs using AI agents on Gumloop, Claude, or Cursor.
Try Extend now
Type what you want done. Sign in and run it live with an AI agent.
|
Extend
Installation
Get StartedCreate a Gumloop Account
To use this MCP, you need a Gumloop account. If you don't have one yet, you can create one for free.
Copy Your Server URL
Copy your MCP server URL and add it to your client. You'll be prompted to authorize on first use.
Tools (10)
List Workflow Runs
List workflow runs with details about execution status, timing, and usage. Supports filtering and automatic pagination.
Get Workflow Run
Get complete details of a specific workflow run including status, outputs, step runs, and usage information.
Run Workflow
Run a workflow with files. Creates workflow runs for each file processed.
Run Processor
Run a processor on a document for extraction, classification, or splitting.
List Processor Runs
List processor runs with filtering and pagination.
Get Processor Run
Get details of a specific processor run including status, outputs, and any edits made during review.
Cancel Processor Run
Cancel a running processor run. Only runs with PROCESSING status can be cancelled.
Delete Processor Run
Permanently delete a processor run and all associated data. This operation cannot be undone.
Parse File
Parse files to get cleaned, chunked content for RAG pipelines, embeddings, or custom processing.
Get Parser Run
Retrieve status and results of a parser run. Use this to poll async parser runs or get completed results.
What is Extend MCP?
The Extend MCP server gives AI agents access to your Extend.ai document processing. Extend.ai is an AI document platform for turning unstructured files into structured, machine-readable output. That means agents can run processors to extract fields from invoices and contracts, classify documents by type, split multi-document files, trigger multi-step document workflows, parse raw files into clean markdown or spatial chunks, and monitor every run. It works across the core Extend.ai objects: workflows, processors, parser runs, and the run records they produce.
If your team pastes invoice fields into spreadsheets by hand, sorts scanned documents into the right buckets, or babysits long document pipelines waiting for results, an AI agent can take over most of that busywork. Describe what you need, and your AI agent will handle the document processing for you.
MCP stands for Model Context Protocol. It’s an open standard that gives AI agents a way to connect to external tools and services. Instead of coding against the Extend.ai API, managing your Bearer token and the pinned API version, and polling run status yourself, you add your Extend.ai API key once. After that, you can run and monitor document processing just by chatting with your AI agent.
Related MCP servers
Reducto MCP Server
Document processing, parsing, extraction, splitting, and editing via Reducto API
Firecrawl MCP Server
Search, scrape, and map websites using Firecrawl's web data API
Google Drive MCP Server
Search and access files in Google Drive
Google Sheets MCP Server
Read, write and update your Google Sheets
Box MCP Server
Interact with Box to search, manage, and access files and folders stored in your Box account
Slack MCP Server
Interact with Slack channels and messages
What you can do with Extend MCP on Gumloop
Extract structured data from documents
Run an extraction processor on a file (provided by public URL or Extend file ID) or on raw text to pull structured fields out of invoices, receipts, contracts, and forms. Your AI agent can override the processor’s default settings per call with a runtime config when one document needs special handling.
Classify and split documents
Run classification processors to label documents by type, and run splitter processors to break a multi-document file into its individual documents. The same run_processor entry point covers extract, classify, and splitter processor types.
Trigger multi-step document workflows
Execute an existing Extend.ai workflow on one or more files or raw text inputs to kick off a full multi-step document pipeline. You can pass pre-computed processor outputs into a workflow step so an agent injects its own results and skips redundant processing, and you can group runs by batch ID.
Parse files into RAG-ready content
Parse a file into cleaned, chunked output as markdown or spatial chunks, built for embeddings, retrieval-augmented generation, and LLM consumption. Choose JSON or URL responses and run it synchronously for real-time needs or asynchronously for larger jobs.
Run synchronously or asynchronously
Processors and parsers can run in sync mode (blocking, up to a five-minute timeout) for immediate results, or in async mode where your agent polls the run for completion. That gives an agent flexibility between real-time lookups and batch processing.
Monitor and inspect runs
List workflow runs and processor runs with filtering by status, ID, batch, processor type, or file name, then pull the full details of any specific workflow run, processor run, or parser run, including outputs, step runs, and usage. Your agent can watch a pipeline and report on what finished, what is still processing, and what needs review.
Cancel and clean up processor runs
Cancel a processor run that is still in progress, or permanently delete a processor run and all of its associated data when you no longer need it. These controls apply to processor runs.
How to connect the Gumloop Extend MCP Server
- 1
Create a free Gumloop account
Sign up at gumloop.com. No credit card required.
- 2
Add the Extend MCP server
Copy your MCP server URL from Gumloop and add it to your preferred client (Claude, Cursor, or Gumloop workflows). You'll authorize on first use.
- 3
Start using Extend in your AI workflows
That's it. Your AI agent can now extract, classify, split, and parse documents, run workflows, and monitor every run. Use it inside a Gumloop automation, in Claude Desktop, or in Cursor.
Extend MCP use cases
Automated invoice data extraction for finance teams
When invoices land, a Gumloop agent can run an Extend.ai extraction processor on each file, pull out vendor, amounts, line items, and dates, and write the structured result into a Google Sheet or your accounting system. AP teams review clean rows instead of retyping fields from PDFs.
Document classification and routing for operations teams
An agent can run a classification processor to label each incoming document by type, then route it to the right destination: contracts to one folder, receipts to another, and anything that returns a needs-review status to a Slack channel for a human to look at. Ops doesn’t have to sort files by hand.
RAG ingestion pipelines for AI and engineering teams
Parse source files into clean markdown or spatial chunks with parse_file, then have an agent embed the chunks and load them into a vector store for retrieval-augmented generation. Engineering teams get document ingestion that an AI agent runs end to end instead of a brittle custom parser.
Contract data extraction for legal and procurement
Run an extraction processor over a batch of contracts to capture parties, renewal dates, payment terms, and obligations, then compile the results into a tracking sheet. An agent can flag contracts that returned a needs-review status so legal focuses only on the ambiguous ones.
Cross-platform document agents
Chain Extend with Google Drive, Google Sheets, Slack, and other MCP integrations in a single agent. Pull a file reference from Drive, run it through an Extend.ai processor or workflow, write the structured output to Sheets, and post a summary to Slack, so your document processing talks to the rest of your stack automatically.
Why use Gumloop for Extend MCP
Your Extend.ai API key, stored securely
Most Extend MCP servers you’ll find on GitHub make you keep your Extend.ai API key in environment variables and write code to handle the Bearer token, the pinned API version, and run polling. With Gumloop you add your Extend.ai API key once, stored securely. No env vars, token management, or coding against the Extend.ai API necessary.
Works with multiple MCP clients
Use the Extend MCP server endpoint in Claude Desktop, Cursor, or directly inside Gumloop agents. Same server URL, works with any MCP client.
Chain Extend with 100+ other integrations
Combine Extend with Google Drive, Google Sheets, Slack, Gmail, and other MCP integrations in a single AI agent. An agent can pull a file from one tool, run it through an Extend.ai processor, and write the structured output anywhere else.
Enterprise-grade and scalable
Built for teams, with role-based permissions and dedicated support for Pro users. The Extend MCP server respects your existing Extend.ai access controls, so an agent only reaches what your connected account can. For details on Gumloop’s security practices, see trust.gumloop.com.
Pricing includes a free plan
You can test the Extend MCP integration on Gumloop’s free tier before committing. Paid plans start at $37/month.
Frequently asked questions
Ship Extend agents in minutes
Connect any AI agent to 100+ MCP servers, fully hosted, zero setup.