MCP Server Js

🚀 YepCode MCP Server Preview

Get Started Now

📚 Project Introduction

The YepCode MCP (Model Context Protocol) Server is a powerful tool that helps you seamlessly connect your AI assistant to your YepCode workspaces and workflows. With the MCP server, you can run code snippets, set environment variables, and execute workflows directly within your AI applications.

✨ Features

• Run Code Snippets: Use the run_code tool to run any code within your AI application.
• Manage Environment Variables: Easily manage environment variables in your workspace using the set_env_var and remove_env_var commands.
• Execute Workflows: Expose your YepCode workflows as MCP tools, allowing your AI assistant to directly invoke these workflows.

📦 Installation

Using the SSE Server

curl https://cloud.yepcode.io/mcp/sk-c2E...RD/sse

Using NPX

Add the following configuration to your package.json:

{
  "scripts": {
    "mcp-server": "yepcode mcp"
  }
}

Then run the command:

npm install yepcode && npm run mcp-server

🔧 Configuration Options

You can enable or disable certain features in the MCP server by setting the YEPCODE_MCP_OPTIONS environment variable. For example:

YEPCODE_MCP_OPTIONS=disableRunCodeTool,skipRunCodeCleanup

Tool Configuration

Run Code Snippets

// Input
{
    "code": string;            // Code to execute
    "language": string;        // Code language (e.g., javascript, python)
    "parameters": object;      // Optional parameters
    "comment": string;         // Execution context comment
}

// Output
{
    "stdout": string;          // Standard output
    "stderr": string;          // Standard error output
    "returnValue": unknown;    // Return value
    "error": string;           // Error message (if any)
}

Set Environment Variables

{
    "key": string,             // Variable name
    "value": string,           // Variable value
    "isSensitive": boolean     // Whether to hide sensitive information (default: true)
}

Remove Environment Variables

{
    "key": string               // Name of the variable to remove
}

Workflow Execution

When you add the mcp-tool tag to your workflow, the MCP server automatically generates a corresponding tool, such as run_ycp_<process_slug>.

Synchronous and Asynchronous Execution

Synchronous Execution

{
    "parameters": any,          // Workflow input parameters
    "options": {               // Optional configuration
        "tag": string,           // Workflow version to execute
        "comment": string        // Execution context comment
    },
    "synchronousExecution": boolean // Whether to wait for execution to complete (default: true)
}

// Response
{
    "executionId": string,     // Unique execution identifier
    "logs": string[],         // Execution logs
    "returnValue": unknown,   // Workflow output
    "error": string           // Error message (if any)
}

Asynchronous Execution

{
    "executionId": string       // Unique execution identifier
}

Get Execution Results

{
    "executionId": string         // Execution ID to query
}

📄 License

This project is licensed under the MIT License. You are free to use, modify, and distribute the code. For more information, please refer to the LICENSE file.

Let's explore the infinite possibilities of AI and automation together!