Gong Tool MCP Server for AI Agent Integration - n8n Workflow

AI Developers building LangChain or similar tool-using agents.
Technical users looking for advanced n8n templates for tool exposure.
Sales Operations teams needing AI access to conversational intelligence data (Gong).
Anyone utilizing the n8n platform to build complex, self-serving API servers.

The 'Gong Tool MCP Server' is a powerful n8n workflow designed to expose specific Gong API operations as tools consumable by AI agents. This automation leverages the Tool Management Control Plane (MCP) framework within n8n, turning standard n8n node functionality into dynamic, queryable actions. This workflow eliminates the need for hardcoding API calls in your AI models; instead, the AI agent decides which operation to run (e.g., 'Get call' or 'Get many users') and automatically populates the required parameters using the specialized $fromAI() expression found in every Gong Tool n8n node.

This setup provides a highly efficient and scalable backend for AI systems. By using this n8n workflow template, you get zero-configuration access to four essential Gong operations, all wrapped in native n8n error handling and response formatting. This is an essential n8n workflow for integrating proprietary data sources with cutting-edge AI logic.

This n8n workflow operates as a specialized server that waits for inbound requests from an AI agent or language model. The logical flow is as follows:

1. MCP Trigger Activation: The workflow starts with the specialized MCP n8n trigger node, which provides a unique webhook URL. When an AI agent determines it needs to perform a Gong operation, it sends a request to this endpoint, specifying the desired tool name and associated parameters.


2. Tool Selection: Based on the AI agent's request, the MCP trigger intelligently routes the execution to the corresponding Gong Tool n8n node (e.g., 'Get many calls').


3. Dynamic Parameterization: Each Gong Tool n8n node uses the expression ={{ $fromAI(...) }} for parameters like limit or call ID. This ensures the parameter values are pulled directly from the input provided by the AI agent, making the request dynamic.


4. Gong Interaction: The selected Gong Tool n8n node executes the API call using the supplied credentials and parameters.


5. Response Handling: The result of the operation (the Gong data) is passed back through the MCP trigger node, formatted, and returned to the querying AI agent, completing the task within the n8n workflow.

To utilize this n8n workflow template effectively, follow these steps:

1. Import the Workflow: Copy the provided JSON data and import it directly into your n8n instance via the 'Workflows' menu, selecting the 'New' button and pasting the JSON code.


2. Add Gong Credentials: Locate any Gong Tool n8n node (e.g., 'Get call'). Open its configuration panel and add or select your Gong API credentials. Ensure this credential setup is completed for the n8n node.


3. Activate the Workflow: Set the n8n workflow to 'Active'.


4. Get the MCP URL: Click on the 'Gong Tool MCP Server' n8n trigger node. Copy the generated webhook URL. This URL is your MCP server endpoint.


5. Connect to AI Agent: Use this MCP URL in your AI agent configuration (e.g., within a LangChain agent setup) to enable it to access these Gong operations as available tools. This n8n workflow is now ready for production use.

This n8n workflow relies on core LangChain and Gong integration nodes:

Gong Tool MCP Server (MCP Trigger):
Function: Acts as the primary entry point and router for the AI tool requests. This is a specialized n8n trigger that exposes the workflow as a unified tool endpoint.
Key Configuration: Path is set to /gong-tool-mcp.
Get call (Gong Tool n8n Node):
Function: Fetches a single call record from Gong.
Key Configuration: Operation is get. The Call ID is dynamically supplied via ={{ $fromAI('Call', , 'string') }}.
Get many calls (Gong Tool n8n Node):
Function: Retrieves multiple call records, allowing for filtering or limiting.
Key Configuration: Operation is getAll. The Limit and Return All parameters are dynamically controlled by the AI agent using $fromAI() expressions.
Get user (Gong Tool n8n Node):
Function: Fetches a single user record from Gong.
Key Configuration: Resource is set to user. User ID is dynamically supplied via ={{ $fromAI('User', , 'string') }}.
Get many users (Gong Tool n8n Node):
Function: Retrieves a list of users.
* Key Configuration: Resource is set to user, Operation is getAll. This n8n node uses $fromAI() for Limit and Return All settings.