The Salesforce MCP Server by Advancedcommunities is a Model Context Protocol server designed for AI assistants to interact with Salesforce organizations. Built with JavaScript/TypeScript and running on Node.js v18 or higher, it offers a direct connection between AI clients and Salesforce. This free mcp category tool enables AI to perform various operations, from data queries to code generation, making Salesforce data and functions accessible to compatible AI clients. It supports the stdio transport protocol, ensuring integration with several popular AI development environments.

What does Salesforce MCP Server connect to?

The Salesforce MCP Server connects AI assistants directly to Salesforce organizations. It functions as a Model Context Protocol server, enabling various AI clients to send commands and receive data from Salesforce instances. The server's architecture allows for interaction with Salesforce environments where an authenticated organization is already configured via the Salesforce CLI. This setup provides a crucial bridge for AI clients to access, query, and manipulate Salesforce data, as well as execute code within the platform.

Core features and tools

The Salesforce MCP Server exposes a specific set of callable tools for AI interaction with Salesforce. These include: List Connected Salesforce Orgs, List Objects in an Org, Describe an Object, Execute Apex Code, Query Records, Query Records to File, Run Apex Tests, Get Apex Test Results, Get Apex Code Coverage, Generate Apex Class, Generate Apex Trigger, List Apex Logs, Get Apex Log, Login to Salesforce Org, Get Server Permissions, Create Record, and install_skill.

Who is it for?

This server is intended for AI assistants and developers who use AI clients to interact with Salesforce organizations. It is particularly useful for those building or using AI clients such as Claude Desktop, Claude Code, Codex CLI, VS Code Copilot, and Cursor, all of which are confirmed compatible. The server provides the necessary interface for these AI tools to perform various operations within Salesforce environments. Skip this if you do not use AI clients or do not need AI to interact with Salesforce organizations.

How to get started

Access is granted through a standard OAuth authorization flow, using an already authenticated Salesforce org via the Salesforce CLI. The server supports the stdio transport protocol for communication with compatible AI clients, ensuring direct interaction.

Use cases

• Executing Apex code within a Salesforce organization using AI commands.
• Querying Salesforce data to retrieve records or object information through AI clients.
• Generating Salesforce metadata, such as new Apex classes or Apex triggers, with AI assistance.

The Salesforce MCP Server offers a direct pathway for AI clients to engage with Salesforce organizations. It provides a defined set of tools for tasks like code execution, data querying, and metadata generation. The server's reliance on existing Salesforce CLI authentication simplifies setup for developers already working with Salesforce. Try generating an Apex Class using one of the compatible AI clients to experience its capabilities.