Nex MCP Server: 47 Tools to Give Claude Desktop, ChatGPT, Cursor, and Windsurf Organizational Context

The Nex MCP Server exposes 47 tools that let any MCP-compatible AI client query and interact with your organization's context graph. It works with Claude Desktop, ChatGPT Desktop, Cursor, Windsurf, and any other tool that supports the Model Context Protocol. Once connected, your AI assistant can search customer records, access deal history, read team communications, manage relationships, and more, all through structured tool calls.

Supported AI Clients

The Nex MCP Server uses stdio transport by default and can be configured for HTTP transport. For Claude Desktop, add the server to your claude_desktop_config.json with your NEX_API_KEY. For Cursor, add it to your .cursor/mcp.json settings. For ChatGPT Desktop, configure it in the ChatGPT settings under the MCP servers section. For Windsurf, add it to your windsurf MCP configuration. Each client connects to the same Nex server and gets access to the same 47 tools and organizational data.

Available Tool Categories

The 47 tools are organized into categories. Context tools (5) let you ask natural language questions and get structured answers from your organizational data. Search tools (1) provide full-text search across all connected sources. Schema tools (8) let you inspect and customize your data model. Record tools manage CRM records for people, companies, and custom entities. Memory tools store and retrieve agent-specific memories. Note tools handle notes attached to records. Task tools manage tasks and follow-ups. Relationship tools map connections between entities. Insight tools surface AI-generated observations. List tools manage collections and segments.

Getting Started

Clone the nex-as-a-skill repository from github.com/nex-crm/nex-as-a-skill, navigate to the mcp directory, and run npm install followed by npm run build. Set your NEX_API_KEY environment variable. Then add the server configuration to your preferred AI client. The server supports both stdio transport for local use and HTTP transport for remote deployments. Full setup instructions and client-specific configuration examples are available in the repository README.