Merge

Provides integration between Merge API and MCP clients, enabling natural-language queries to manage Merge data and workflows.

python

16

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "merge-api-merge-mcp": {
      "command": "uvx",
      "args": [
        "merge-mcp"
      ],
      "env": {
        "MERGE_TENANT": "US",
        "MERGE_API_KEY": "<your_api_key>",
        "MERGE_ACCOUNT_TOKEN": "<your_account_token>"
      }
    }
  }
}

This MCP server enables natural-language interaction with your Merge API data via MCP-compatible clients. You can query entities, learn about your data models, and create or update records through conversational prompts, using a unified interface across multiple Merge API categories.

How to use

Connect to the Merge MCP server from your MCP client and start interacting with your Merge data using natural language. You can ask questions about entities, list and inspect data models, and perform create or update operations through conversational flows. The server exposes tools based on your enabled scopes, and it adapts to the Merge API categories you have access to.

How to install

Prerequisites: you need a Merge API key and account token, Python 3.10 or higher, and the uv runtime. Install uv first or use the Python package manager to install it.

curl -LsSf https://astral.sh/uv/install.sh | sh

pip install uv

Alternatively, install uv via pipx:

pipx install uv

Configuration and running the MCP server

Use the provided MCP configuration to run the server locally through uvx. This configuration sets the command to run and the environment variables needed for your Merge API access.

{
  "mcpServers": {
    "merge_mcp": {
      "command": "uvx",
      "args": ["merge-mcp"],
      "env": {
        "MERGE_API_KEY": "your_api_key",
        "MERGE_ACCOUNT_TOKEN": "your_account_token"
      }
    }
  }
}

Additional setup notes

If you are using a client like Claude Desktop or a custom Python client, you can adapt the same configuration to connect. Ensure you replace your API keys with real values and point to the correct command path for uvx on your system.

Security and scopes

Scopes control which Merge API categories and operations are available to the MCP server by enabling specific models and permissions on your Linked Account. When you start the server you can specify a subset of scopes with a --scopes option to limit access.

Troubleshooting and notes

If tools do not appear in your client after connecting, verify that your MERGE_API_KEY and MERGE_ACCOUNT_TOKEN are correct and that the uvx command is accessible on your system. Restart your client after updating tokens or the uvx path.

Examples of common tasks

Ask questions about data models to learn relationships between entities, list records within an entity, create new records, or update existing ones using the same conversational interface.