home / mcp / honeycomb mcp server

Honeycomb MCP Server

Honeycomb MCP (Model-Controller-Presenter) Server implementation

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "kajirita2002-honeycomb-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "@kajirita2002/honeycomb-mcp-server"
      ],
      "env": {
        "HONEYCOMB_API_KEY": "your_honeycomb_api_key"
      }
    }
  }
}

This Honeycomb MCP Server lets Claude AI interact with the Honeycomb API through the Model Context Protocol, enabling you to retrieve, create, and manage datasets, queries, events, boards, and more directly from your Claude workflows.

How to use

You integrate the Honeycomb MCP Server into your Claude AI setup and then use tools to manage Honeycomb resources. Start by authenticating with your Honeycomb API key, then list datasets, inspect dataset details, create and run queries, and fetch results. You can also manage dataset definitions, boards, and columns. Each action is exposed as a distinct tool you can call from Claude to perform a specific operation.

How to install

Prerequisites you need before installation are you have Node.js 18 or higher and a Honeycomb API key. Follow these steps to install and run the MCP Server.

Install the MCP server globally so you can run it from anywhere, or use it directly with npx.

# Install globally
npm install -g @kajirita2002/honeycomb-mcp-server

# Or use directly with npx
npx @kajirita2002/honeycomb-mcp-server

Configuration and startup

Configure the MCP server to supply your Honeycomb API key and wire up the server command. The following is an MCP configuration example you can place in your mcp_config.json.

"honeycomb": {
  "command": "npx",
  "args": ["-y", "@kajirita2002/honeycomb-mcp-server"],
  "env": {
    "HONEYCOMB_API_KEY": "your_honeycomb_api_key"
  }
}

Starting the server

Once configured, start the MCP server to begin handling requests from Claude.

npm start

Notes on configuration options

The server can be run using different executable approaches. In addition to starting with npm start, you can run via npx to execute the MCP server directly, which is useful in environments where you prefer an on-demand approach.

Available tools

honeycomb_auth

Authenticates with the Honeycomb API and validates your API key using the provided environment variable.

honeycomb_datasets_list

Lists all datasets available in your Honeycomb environment.

honeycomb_dataset_get

Retrieves detailed information about a specific dataset by its slug.

honeycomb_columns_list

Lists all columns in a specified dataset, with optional filtering by column key.

honeycomb_query_create

Creates a new query for a given dataset, including calculations, time range, and filters.

honeycomb_query_get

Fetches information about a specific query by dataset slug and query ID.

honeycomb_query_result_create

Executes a query and returns the results, with options to adjust series and aggregates.

honeycomb_query_result_get

Retrieves the results of a previously executed query.

honeycomb_dataset_definitions_list

Lists dataset definitions with pagination and sorting support.

honeycomb_boards_list

Lists all available boards in the Honeycomb environment.

honeycomb_board_get

Gets detailed information about a specific board by its ID.