LABSCO SUMMARY

The canvas decides which of your Python functions the agent may call.

The tool list is not fixed by a vendor — it is whichever UDFs you left visible on a canvas, so the agent's reach is edited the same way you edit the pipeline. It also keeps the heavy part in place: the UDF runs on Fused and the model receives the answer, not the dataset.

What it is

Fused's MCP surface: any Canvas in Workbench can be published as an MCP server, exposing the UDFs on it — Python functions that join and process data from files, cloud storage, databases and STAC catalogs — as callable tools.

What you get

A ready-to-run connect command from the canvas itself: open it in Workbench, click Share, then Copy MCP
Every visible UDF on the canvas discoverable and callable as a tool, with its parameters
A visibility toggle per UDF node, so helper functions still run when called by other UDFs but stay out of the agent's tool list
The same MCP URL working in Claude Code, the Claude desktop app, claude.ai, Cursor, VS Code and any other MCP client
Compute that stays behind the endpoint — the agent gets the result, not the dataset
An ai-chat widget you can drop onto the canvas itself, with access to every UDF wired into that widget node

Requirements

A Fused account and a canvas in Workbench. Connection is over HTTP with a canvas token in the URL, added with claude mcp add --transport http. On Team and Enterprise plans only Owners can add the connector org-wide, after which each member connects and authenticates individually; the ai-chat widget requires viewers to be logged in to fused.io.

Setup effort

Build from source — clone the repository, then follow the project's own run instructions

See installation options

USE CASES

01

Letting an agent query a dataset too big to hand it

Wrap the query in a UDF, expose the canvas, and the agent gets rows back while the compute stays on Fused.

02

Serving the data your team refreshes every day

Point a UDF at wherever the automation writes, and any agent on any machine reads the current version through the same endpoint.

03

Keeping helper functions out of the agent's way

Hide the internal UDF nodes in Workbench — they still run when other UDFs call them, they just stop appearing as tools.

04

Putting a chat next to the data instead of in another window

Add the ai-chat widget to the canvas and wire the UDF nodes into it, so questions get answered where the pipeline lives.

BEFORE YOU INSTALL

What to know first

Requires

Python 3.11

INSTALLATION

Connect your client

Server setup

TERMINAL

git clone https://github.com/fusedio/fused-mcp.git

MCP Clients

Claude Desktop

CLAUDE DESKTOP · JSON

{
  "mcpServers": {
    "qgis": {
      "command": "/Users/<YOUR_USERNAME>/.local/bin/uv",
      "args": [
        "<YOUR_ARGS>"
      ]
    }
  }
}

Bundled Python client — Linux

BUNDLED PYTHON CLIENT — LINUX · TERMINAL

uv run main.py --agent get_current_time
# in another terminal:
uv run client.py http://localhost:8080/sse