Easily launch the Glif MCP server and toolkits in a Meshagent Room in the cloud:

​

Environment Variables

Set the following environment variables as needed:

• GLIF_IDS: Comma-separated list of Glif IDs (leave blank for all).
• IGNORE_SAVED_GLIFS: Set to false to include saved Glifs (default: “false”).
• GLIF_API_TOKEN: Your Glif API token (replace your-token-here).

​

Steps

1. Install the Meshagent package:

   Copy

   Ask AI

   pip install "meshagent[all]"
   

2. Sign up and authenticate:

   • Read the Meshagent CLI getting started docs and obtain an auth token.

3. Start the MCP Glif server as an agent inside a Meshagent Room:

   Copy

   Ask AI

   meshagent service test --room=test --role=agent --image=meshagent/mcp-glif:latest \
       --env MESHAGENT_PORT=8001 \
       --env GLIF_IDS=<your-glif-ids> \
       --env GLIF_API_TOKEN=<your-api-token> \
       --env IGNORE_SAVED_GLIFS=false \
       --port="num=8001 path=/webhook liveness=/ type=meshagent.callable" \
       --name=mcp-glif-service
   

   • This command launches the MCP server in a Meshagent room named test. The room is ephemeral and will close if it goes inactive.

4. Join the Room with a Meshagent-powered chatbot and toolkits:

   Copy

   Ask AI

   meshagent chatbot join --room=test --agent-name=sample --name=sample --toolkit=mcp-glif
   

   • This command starts a chatbot agent in your Meshagent room using the Glif toolkit (mcp-glif).
   • Multiple toolkits and agents can co-exist in the same room.
   • The output of this command provides a direct link to your room for live interactions with agents and tools.

​

Tools Available

Easily run glif.app AI workflows inside your LLM: image generators, memes, selfies, and more. Glif supports all major multimedia AI models inside one app.

What is an MCP Server?

​

Tools Provided by this Server

​

Tool Details

​

glif_info

Get detailed information about a glif including input fields
Parameters:

• id (string): The ID of the glif to show details for

​

list_featured_glifs

Get a curated list of featured glifs

​

list_saved_glif_tools

List all saved glif tools

​

my_glif_user_info

Get detailed information about your user account, recent glifs, and recent runs

​

my_glifs

Get a list of your glifs

​

remove_all_glif_tools

Remove all saved glif tools and return to a pristine state

​

remove_glif_tool

Remove a saved glif tool
Parameters:

• toolName (string): The tool name of the saved glif to remove

​

run_glif

Run a glif with the specified ID and inputs
Parameters:

• id (string): The ID of the glif to run
• inputs (array): Array of input values for the glif

​

save_glif_as_tool

Save a glif as a custom tool
Parameters:

• id (string): The ID of the glif to save
• toolName (string): A unique tool name
• description (string, optional): Custom description (default: glif description)
• name (string, optional): Custom name for the tool (default: glif name)

​

search_glifs

Search for glifs by query string
Parameters:

• query (string): Search query string

​

Meshagent & Glif Links

• Meshagent: https://www.meshagent.com/
• Meshagent Getting Started Docs: https://docs.meshagent.com/room_api/get_started
• Meshagent API Overview: https://docs.meshagent.com/room_api/overview
• Meshagent MCP-GLIF Docker Image: https://hub.docker.com/r/meshagent/mcp-glif

For more on running AI apps with the Model Context Protocol: Why is it safer to run MCP Servers with Docker?