glif
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
claude mcp add --transport stdio glifxyz-glif-mcp-server npx -y @glifxyz/glif-mcp-server@latest \ --env GLIF_API_TOKEN="your-token-here"
How to use
The glif MCP server integrates Glif workflows into your preferred AI assistant setup. It exposes a set of Core Tools that let you run workflows, fetch detailed information about workflows, and manage your discovery of available workflows. There are optional Discovery Tools for searching and listing workflows, as well as Agent Tools to inspect and load agents for enhanced automation. To get started, obtain a Glif API token and configure the MCP server in your client (Claude or similar) so it can call into the Glif API through the server. The Core Tools are designed for direct workflow execution and information retrieval, while the Discovery Tools help you explore what’s available and your own workflows. If you enable Agent Tools, you can also load agents and access their personalities and associated workflows.
How to install
Prerequisites:
- Node.js installed (for the npx option) or a local checkout to run a built server via Node.js
- Access to a Glif API token from glif.app/settings/api-tokens
Option A: Run via npx (recommended)
- Install Node.js if you don’t have it.
- Get your Glif API token from https://glif.app/settings/api-tokens
- Add the MCP server to your client config (e.g., Claude) with: { "mcpServers": { "glif": { "command": "npx", "args": ["-y", "@glifxyz/glif-mcp-server@latest"], "env": { "GLIF_API_TOKEN": "your-token-here" } } } }
- Save the config and start using the glif tools through your MCP client.
Option B: Run from a local checkout
- Clone the repository: git clone https://github.com/glifxyz/glif-mcp-server
- Install dependencies and build: cd glif-mcp-server npm install npm run build
- Configure your MCP client to point at the built server, for example: { "mcpServers": { "glif": { "command": "node", "args": ["/path/to/glif-mcp/build/index.js"], "env": { "GLIF_API_TOKEN": "your-token-here" } } } }
- You can optionally preload specific workflow IDs via GLIF_IDS in the environment to auto-load tools.
Note: Replace "/path/to/glif-mcp-build/index.js" with the actual path to your built server file. The Glif API token must be kept secret and not exposed in client logs.
Additional notes
Environment variables and tips:
- GLIF_API_TOKEN is required to authenticate with Glif APIs.
- GLIF_IDS (optional) should be a comma-separated list of workflow IDs to preload as tools (e.g., cm2v9aiga00008vfqdiximl2m,cm2v98jk6000r11afslqvooil).
- IGNORE_DISCOVERY_TOOLS can disable discovery tools if set to true.
- AGENT_TOOLS can enable agent tools; by default agents are disabled.
- You can load workflows by ID or discover via list_featured_workflows and search_workflows.
- If you encounter environment variable issues, ensure the hosting environment passes the variables to the MCP process correctly and that there are no conflicting tokens.
- When using the local build, remember to rebuild after changes to src/index.ts as part of the release cycle.
Related MCP Servers
n8n
A MCP for Claude Desktop / Claude Code / Windsurf / Cursor to build n8n workflows for you
mcp-router
A Unified MCP Server Management App (MCP Manager).
conductor-tasks
A task management system designed for AI development
dev-to
A remote Model Context Protocol (MCP) server for interacting with the dev.to public API without requiring authentication.
ToolRAG
Unlimited LLM tools, zero context penalties — ToolRAG serves exactly the LLM tools your user-query demands.
google-scholar
An MCP server for Google Scholar written in TypeScript with Streamable HTTP