ShotGrid

ShotGrid MCP Server provides a high-performance bridge that lets AI assistants access Autodesk ShotGrid data directly. It exposes 40+ tools for creating, querying, and updating ShotGrid entities and media, enabling seamless interactions between AI copilots and production workflows.

How to use

You connect your MCP client to the ShotGrid MCP Server to start asking questions or issuing actions that operate on ShotGrid data. Start with a local, stdio-based setup for desktop assistants, then switch to HTTP for remote access when you need to reach the server from another machine or environment. The server handles authentication with ShotGrid and exposes a rich set of tools for common production tasks.

How to install

Prerequisites: you need Python and a compatible runtime environment. You will install the MCP server package, then run it using the local stdio transport or expose it via HTTP for remote clients.

# Using uvx (recommended for local stdio transport)
uvx pip install shotgrid-mcp-server

# Or using pip directly
pip install shotgrid-mcp-server

Configuration and startup

Set your ShotGrid credentials in the environment so the server can authenticate with ShotGrid.

export SHOTGRID_URL="https://your-site.shotgunstudio.com"
export SHOTGRID_SCRIPT_NAME="your_script_name"
export SHOTGRID_SCRIPT_KEY="your_script_key"

Start the server using the stdio transport for local development or testing with Claude Desktop, Cursor, or other MCP clients.

uvx shotgrid-mcp-server

To enable remote HTTP access, expose the server on the desired host and port.

uvx shotgrid-mcp-server http --host 0.0.0.0 --port 8000

MCP client configuration examples

Configure your MCP clients to connect to the local stdio server or the HTTP endpoint. The following examples show how to configure clients like Claude Desktop, Cursor, and VS Code to reach the ShotGrid MCP Server.

{
  "mcpServers": {
    "shotgrid": {
      "command": "uvx",
      "args": ["shotgrid-mcp-server"],
      "env": {
        "SHOTGRID_URL": "https://your-site.shotgunstudio.com",
        "SHOTGRID_SCRIPT_NAME": "your_script_name",
        "SHOTGRID_SCRIPT_KEY": "your_script_key"
      }
    }
  }
}

Deployment considerations

You can deploy the server via FastMCP Cloud, ASGI servers, or Docker for production. If you choose an ASGI deployment, you can use a standard Python web server to run the application.

# ASGI deployment example
uvicorn shotgrid_mcp_server.asgi:app --host 0.0.0.0 --port 8000

Troubleshooting and tips

Verify that your ShotGrid credentials are valid and that the server can reach the ShotGrid API. Check network accessibility, environment variable names, and port bindings if you cannot connect from an MCP client.

Example prompts for your AI assistant

Once connected, you can ask for common production tasks such as querying shots, managing playlists, or adding notes to shots to keep teammates aligned.

Available tools

create_entity

Create a new ShotGrid entity with specified fields.

find_one_entity

Retrieve a single entity by identifier or filter.

search_entities

Query multiple entities matching criteria.

update_entity

Update fields on an existing entity.

delete_entity

Remove an entity from ShotGrid.

batch_create

Create multiple entities in a single operation.

batch_update

Update multiple entities in one call.

batch_delete

Delete multiple entities in a single call.

download_thumbnail

Download a thumbnail for a media asset.

upload_thumbnail

Upload or attach a thumbnail to a media asset.

shotgrid.note.create

Create a note on a ShotGrid entity.

shotgrid.note.read

Read notes attached to a ShotGrid entity.

shotgrid.note.update

Update an existing note.

create_playlist

Create a playlist to organize related shots or assets.

find_playlists

Find and retrieve playlists.

sg.find

Direct access to ShotGrid search API.

sg.create

Direct access to ShotGrid create API.

sg.update

Direct access to ShotGrid update API.

sg.batch

Direct access to ShotGrid batch API.