home / mcp / g-gremlin-salesforce mcp server

g-gremlin-salesforce MCP Server

Provides a dedicated MCP launcher that connects Salesforce tools to MCP clients, delegating read-only and write-enabled modes.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "mikeheilmann1024-g-gremlin-salesforce-mcp": {
      "command": "g-gremlin-sfdc-mcp",
      "args": []
    }
  }
}

You can run a dedicated MCP launcher that connects Salesforce tooling to MCP clients without invoking the broader CLI. This server exposes read-only analysis tools by default and can opt-in to expose write-enabled tools, allowing you to manage Salesforce-related operations through MCP workflows.

How to use

You use this MCP server by configuring your MCP client to connect to the Salesforce-focused MCP launcher. The server delegates tool access to either read/analyze capabilities or, when enabled, write-enabled tools. Your client can request read-only tool information by default and switch to writes when you enable them in the server configuration.

How to install

Prerequisites: you need Python and pip, and you should have a Python environment available. You will also use a local MCP launcher to expose the Salesforce tools to your MCP clients.

Install the Salesforce MCP launcher components with your preferred Python package manager.

pip install g-gremlin
pip install g-gremlin-sfdc-mcp

# Start or configure your MCP client after installation

# Example usage depends on your MCP client setup; see your client’s documentation for connection steps.

Additional notes and configuration

To expose write-enabled tools, you will configure the MCP client to pass the flag that enables writes to the Salesforce MCP launcher. When writes are enabled, apply calls may require a plan_hash to ensure safe operations.

Health checks and basic verification are supported to confirm the MCP launcher is reachable and responding to requests.

Health check

g-gremlin-sfdc-mcp --check

Development

If you are developing or testing locally, clone the repository, install development dependencies, and run tests to validate behavior.

git clone https://github.com/mikeheilmann1024/g-gremlin-salesforce-mcp
cd g-gremlin-salesforce-mcp
pip install -e ".[dev]"
pytest

Security considerations

When enabling write access, ensure you apply proper authorization and audit controls. Write operations can modify Salesforce tooling, so restrict access to trusted clients and monitor tool usage.

Available tools

serve

Launches the MCP server to expose Salesforce tooling for read/analyze operations to MCP clients.

enable_writes

Enables exposure of write-enabled tools, allowing write operations through MCP with appropriate safeguards.