Unleash MCP server

The Unleash Feature Flag MCP Server allows AI agents to manage feature flags through the Unleash API, implementing the Model Context Protocol (MCP) to enable programmatic feature flag management.

Installation

You can install the package globally from npm:

npm install -g @ylin6/unleash-ff-mcp-server

Alternatively, run it directly using npx without installation:

npx @ylin6/unleash-ff-mcp-server

Configuration

Before using the server, you need to set the following environment variables:

• UNLEASH_API_URL: The URL of your Unleash API instance
• UNLEASH_AUTH_TOKEN: The authentication token for your Unleash instance

Available Tools

Get Projects

Retrieves a list of all projects in the Unleash instance.

Get Features

Retrieves all feature flags within a specific project.

Parameters:

• projectId: The ID of the project

Create Feature Flag

Creates a new feature flag within a project.

Parameters:

• projectId: The ID of the project
• name: The name of the feature flag
• description: A description of the feature flag
• type: The type of the feature flag (e.g., "release", "experiment", "operational", "kill-switch")

Update Feature Flag

Updates an existing feature flag.

Parameters:

• projectId: The ID of the project
• featureId: The ID of the feature flag
• description: A new description for the feature flag
• type: A new type for the feature flag

Get Feature Flag

Retrieves details about a specific feature flag.

Parameters:

• projectId: The ID of the project
• featureId: The ID of the feature flag

Using with Cursor

To use this MCP server with Cursor, add the following command in your Cursor settings:

env UNLEASH_API_URL=XXXX UNLEASH_AUTH_TOKEN=XXX npx -y @ylin6/unleash-ff-mcp-server

Replace XXXX and XXX with your actual Unleash API URL and authentication token.

Example Usage

Here's an example conversation with an AI assistant like Claude in Cursor:

You: Show me all the feature flags in the 'dashboard' project

Claude: I'll fetch all the feature flags in the 'dashboard' project for you.
[Claude uses the getFeatures tool with projectId='dashboard']

Claude: Here are all the feature flags in the 'website' project:
- new-homepage (type: release)
- dark-mode (type: experiment)
- beta-footer (type: operational)
...

How to add this MCP server to Cursor

There are two ways to add an MCP server to Cursor. The most common way is to add the server globally in the ~/.cursor/mcp.json file so that it is available in all of your projects.

If you only need the server in a single project, you can add it to the project instead by creating or adding it to the .cursor/mcp.json file.

Adding an MCP server to Cursor globally

To add a global MCP server go to Cursor Settings > MCP and click "Add new global MCP server".

When you click that button the ~/.cursor/mcp.json file will be opened and you can add your server like this:

{
    "mcpServers": {
        "cursor-rules-mcp": {
            "command": "npx",
            "args": [
                "-y",
                "cursor-rules-mcp"
            ]
        }
    }
}

Adding an MCP server to a project

To add an MCP server to a project you can create a new .cursor/mcp.json file or add it to the existing one. This will look exactly the same as the global MCP server example above.

How to use the MCP server

Once the server is installed, you might need to head back to Settings > MCP and click the refresh button.

The Cursor agent will then be able to see the available tools the added MCP server has available and will call them when it needs to.

You can also explictly ask the agent to use the tool by mentioning the tool name and describing what the function does.