sf-mcp-server
If you are the rightful owner of sf-mcp-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.
This repository demonstrates the integration between a Salesforce org and an LLM agent using the Model Context Protocol (MCP).
The Salesforce MCP Server is designed to facilitate seamless communication between a Salesforce organization and a language model agent, specifically leveraging the Model Context Protocol (MCP). The server is implemented using FastMCP and runs locally, exposing core Salesforce REST operations as MCP tools and resources. The agent, powered by GPT-4, can autodiscover these tools and utilize them as needed. This setup allows for efficient data handling and operations within Salesforce, driven by advanced language model capabilities. The repository includes necessary scripts and configurations to set up a connected app in Salesforce, manage environment variables, and deploy the server for production use.
Features
- End-to-end integration between Salesforce and LLM using MCP.
- FastMCP server exposing Salesforce REST operations.
- GPT-4 powered agent for autodiscovery and tool utilization.
- Support for OAuth 2.0 and secure token management.
- Production deployment guidance with Docker and API Gateway.
Tools
server.py`
FastMCP server, providing MCP interface for Salesforce REST operation
agent.py`
4 Driver agents, automatic discovery and calling MCP tools
