Document processing in Cursor

Connect the Extend MCP server to Cursor and parse, extract, classify, split, and edit documents from the agent panel.

Step 1

Connect the MCP server

Add the server to ~/.cursor/mcp.json, or to a project's .cursor/mcp.json to scope it to one repo.
Complete the OAuth sign-in when prompted and choose the workspaces the client may use and the environment it acts in.
Ask the Cursor agent to process a document; it picks the right Extend tool.

Details on the MCP page Cursor setup in the docs

~/.cursor/mcp.json

{
  "mcpServers": {
    "extend": {
      "url": "https://mcp.extend.ai/mcp"
    }
  }
}

Step 2

Install the skills

Skills teach the agent how to drive Extend correctly. The CLI skill is generated from the extend command tree; the API skill is a brief on the API, SDKs, and schema rules.

extend-cli

curl -fsSL https://extend.ai/install.sh | sh && extend setup

Installs the extend command and writes the extend-cli skill to ~/.agents/skills/extend-cli/SKILL.md.

extend-api

npx skills add extend-hq/extend-agent-plugin --skill extend-api

Installs the extend-api context file at the project level so it can be committed with your repo.

Try these prompts

Copy one into the agent and let it pick the tools.

Verify it works

The agent calls get_me and lists the workspaces and environments it was granted. No credits are spent.

Frequently asked questions

Reconnect the server in your client to re-run the OAuth sign-in. Most clients have a reconnect or re-authenticate action on the server's settings entry; disconnecting and re-adding also works.

It's usually targeting the wrong place: resources are separated per workspace and per Test/Production environment. Ask the agent to check its granted workspaces and environments with the get_me tool and to target where the resource actually lives.