YepCode MCP Server is an MCP (Model Context Protocol) server that enables AI platforms to interact with YepCode's infrastructure by turning LLM-generated scripts and YepCode processes into AI-ready tools. It supports executing code in YepCode's secure, isolated environments and exposing processes as individual MCP tools that AI assistants can call directly. The server can be deployed remotely via a hosted SSE server or locally via NPX or Docker, with authentication via YepCode API credentials. Clients connect using either an SSE URL with a token or an HTTP endpoint with an Authorization header. It provides real-time process control and enterprise-grade security, along with environment variable management (set_env_var, remove_env_var) and access to an built-in storage system (list, upload, download, delete). Tool exposure is governed by YEPCODE_MCP_OPTIONS and YEPCODE_MCP_TOOLS to control code execution capabilities and which processes are surfaced as tools. Debugging is supported via the MCP Inspector, and a full API management suite (yc_api, yc_api_full) enables comprehensive control over processes, schedules, variables, storage, executions, and modules. This design ensures universal compatibility with any AI platform that supports MCP.