Get the FREE Ultimate OpenClaw Setup Guide →

ClimateTraceKMP

Kotlin/Compose Multiplatform project to show climate related emission data from https://climatetrace.org/data.

Installation
Run this command in your terminal to add the MCP server to Claude Code.
Run in terminal:
Command
claude mcp add --transport stdio joreilly-climatetracekmp java -jar /path/to/climatetracekmp/mcp-server/build/libs/serverAll.jar --stdio \
  --env JAVA_OPTS="Optional JVM options (e.g., -Xmx2g)"

How to use

This MCP server exposes tools via the Kotlin MCP SDK to provide per-country emission data derived from Climate TRACE data. It’s built as part of a Kotlin Multiplatform project and is wired into the mcp-server module so you can plug it into Claude Desktop or other MCP clients by starting the packaged jar and loading the server configuration. The server acts as a simple endpoint that returns country-level emission data, enabling downstream tooling and UI clients to fetch up-to-date climate metrics.

How to install

Prerequisites:

  • Java (JDK 11+)
  • Gradle (wrapper is fine)
  1. Clone the repository and navigate to the mcp-server module: git clone https://github.com/joreilly/ClimateTraceKMP.git cd ClimateTraceKMP/mcp-server

  2. Build the MCP server artifact (uses the Kotlin MCP SDK and Gradle Shadow to produce a runnable jar): ./gradlew shadowJar

  3. Verify the artifact exists: ls build/libs/

    you should see serverAll.jar among the outputs

  4. Run the MCP server locally (adjust the jar path to where your build produced the jar): java -jar build/libs/serverAll.jar --stdio

  5. Add to your MCP client config (examples shown in the repository README show how to wire this into Claude Desktop). The key is to point to the jar and include the --stdio flag so the MCP framework can communicate via standard I/O.

Additional notes

Tips and common considerations:

  • Ensure the path to serverAll.jar matches your build output location.
  • If you need more memory for the JVM, you can set JAVA_OPTS (e.g., -Xmx4g).
  • The server is intended to be wired via the MCP configuration as shown in the README example; it responds with per-country emission data suitable for integration with clients like Claude Desktop.
  • If you’re integrating with an IDE or CI, you can adjust the Gradle task to produce a standalone jar and then reference that in mcp_config.

Related MCP Servers

Sponsor this space

Reach thousands of developers