home / mcp / ambivo claude mcp server

Ambivo Claude MCP Server

Provides Claude-based natural language access to Ambivo CRM data via entity queries.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "ambivo-corp-ambivo-mcp-server": {
      "command": "ambivo-mcp-server",
      "args": []
    }
  }
}

Ambivo Claude MCP Server enables you to interact with Ambivo CRM data using Claude AI through natural language queries. It authenticates with JWT tokens, handles errors gracefully, and supports reliable retry and caching to provide a smooth querying experience against your entity data.

How to use

You use this MCP server with an MCP client to authenticate, then issue natural language queries against Ambivo entity data. Start by configuring authentication, then perform natural language queries to retrieve or filter leads, contacts, opportunities, and more. You will receive results in formats that suit your display needs, such as tables, natural text, or a combination of both.

Typical usage pattern you’ll follow: set your authentication token once, then run a natural language query describing the data you want. If you need direct structured access, you can also query entity data with precise filters and sorting.

Notes on behavior you can rely on: the server enforces Bearer token authentication, automatically retries transient failures with exponential backoff, and caches token validation to keep requests fast. If an error occurs, you’ll see clear messages indicating authentication problems, parameter validation issues, or network timeouts.

How to install

Prerequisites: you need Python and pip installed on your system.

Install the MCP server package from PyPI.

pip install ambivo-mcp-server

Alternatively, install from source by cloning the repository and installing in editable mode.

git clone https://github.com/ambivo-corp/ambivo-mcp-server.git
cd ambivo-mcp-server
pip install -e .

Run the server directly after installation.

ambivo-mcp-server
```

Or run via the Python module interface.

python -m ambivo_mcp_server.server

Configuration and running notes

You can start with the default configuration and adjust as needed. The server uses a default base URL, timeout, and content type, and you can modify these settings in the client if you require different behavior.

Authentication and endpoints

Set your JWT Bearer token once you start interacting with the server. The token is then included with all subsequent requests. The server validates the token and returns useful errors when authentication fails.

Key endpoints handled by the MCP server include the natural language query endpoint to process queries against Ambivo entity data and an endpoint for direct entity data access.

Error handling and reliability

Expect comprehensive error messages for authentication problems, HTTP errors, validation issues, and network timeouts. The server includes retry logic with exponential backoff to gracefully recover from transient failures.

What you can do next

Set your authentication token, perform natural language queries to retrieve CRM data, and use direct entity queries for structured access. Explore the supported response formats to display results in a way that fits your workflow.

Available tools

set_auth_token

Sets the JWT Bearer token used for authenticating requests to the Ambivo API.

natural_query

Executes natural language queries against Ambivo entity data and returns results in preferred formats.