yepcode/mcp-server-js

An MCP (Model Context Protocol) server that enables AI platforms to interact with YepCode's infrastructure. Run LLM-generated scripts and turn your YepCode processes into powerful tools that AI assistants can use directly. YepCode is the perfect environment to build a **dynamic MCP tools server**: expose each process as a tool (with OAuth, API tokens, or your credentials), define each tool's parameters with **JSON Schema** for full flexibility, and implement tools in **Python** or **Node.js**—all in one server that mixes multiple languages.
Why YepCode MCP Server?
- **Seamless AI Integration**: Convert YepCode processes into AI-ready tools with zero configuration
- **Real-time Process Control**: Enable direct interaction between AI systems and your workflows
- **Enterprise-Grade Security**: Execute code in YepCode's isolated, production-ready environments
- **Universal Compatibility**: Integrate with any AI platform supporting the Model Context Protocol
YepCode: The Perfect Environment for a Dynamic MCP Tools Server
YepCode is built to be the ideal platform for running a **dynamic MCP tools server**:
- **One process, one tool**: Each YepCode process can be exposed as an MCP tool. Tag your processes (e.g. mcp-tool, core, automation) and they become tools that AI assistants can invoke. You can secure access with **OAuth**, **API tokens**, or your existing YepCode credentials—each tool runs in your workspace with the same security model.
- **Full control over tool parameters**: Every tool can define its own **parameter schema as JSON Schema**. You get complete flexibility to describe inputs (types, descriptions, required fields, enums, defaults, etc.), so the AI receives rich metadata and can call your tools correctly.
- **Polyglot tool implementations**: Implement tools in **Python** or **Node.js** (or both). The same MCP server can expose tools backed by different runtimes—think of it as one MCP server that mixes implementations across several languages.
For complete documentation, see the YepCode MCP Server docs.
Installation
This package lets you run the YepCode MCP server **locally** or in your own infrastructure (NPX, Docker, or custom deployment). Integrate it with AI platforms like Cursor or Claude Desktop.
> **Tip:** From your YepCode account you also have access to a hosted MCP server that doesn't require local installation. The connection URL is always: https://cloud.yepcode.io/mcp
Prerequisites
Get your YepCode API credentials:
1. Sign up to YepCode Cloud
2. Visit Settings > API credentials to create a new API token.
Using NPX
Make sure you have Node.js installed (version 18 or higher), and use a configuration similar to the following:
Using Docker
1. Build the container image:
2. Use a configuration similar to the following:
Debugging
Debugging MCP servers can be tricky since they communicate over stdio. To make this easier, we recommend using the MCP Inspector, which you can run with the following command:
This will start a server where you can access debugging tools directly in your browser.
YepCode MCP Tools Reference
The MCP server provides several tools to interact with YepCode's infrastructure:
Code Execution
run_code
Executes code in YepCode's secure environment.
MCP Options
YepCode MCP server supports the following options:
- runCodeCleanup: Skip the run_code cleanup. By default, run_code processes source code is removed after execution. If you want to keep it for audit purposes, you can use this option.
- skipCodingRules: Skip including coding rules in the run_code tool definition. By default, JavaScript and Python coding rules from YepCode documentation are included in the tool schema to guide AI-generated code. If you want to skip this for faster tool initialization or smaller tool definitions, you can use this option.