Get the FREE Ultimate OpenClaw Setup Guide →

catalog-api

MCP server from davidlbl/catalog-api-mcp-server

Installation
Run this command in your terminal to add the MCP server to Claude Code.
Run in terminal:
Command
claude mcp add --transport stdio davidlbl-catalog-api-mcp-server {{JAVA_HOME}} -jar {{LOCAL_PATH}}\\catalog-api-mcp-server\\target\\catalog-api-mcp-server-0.0.1-SNAPSHOT.jar \
  --env JAVA_HOME="Path to Java JDK installation" \
  --env LOCAL_PATH="Path to project workspace containing catalog-api-mcp-server"

How to use

This MCP server (catalog-api-mcp-server) is a lightweight tool to validate and manage the status of the P&C catalog API. It is started from a Java JAR built from the project and exposes a set of predefined tools that can be used to query, validate, and interact with the local catalog API implementation. Once the MCP agent is activated in your editor (Visual Studio Code in this setup), you need to configure the MCP settings to point to the local JAR produced by the catalog-api-mcp-server project. After starting the MCP, you will be prompted to select the agent and you will see the list of tools that were configured. These tools enable you to perform operations such as checking catalog status, validating responses, and performing read operations against the local service as defined in the project (see EcommerceService for the concrete tool definitions). You can then run these tools to interact with the server locally and verify behavior during development.

How to install

Prerequisites:

  • Java JDK (required to run the MCP server JAR)
  • Maven or Gradle to build the catalog-api-mcp-server project
  • Visual Studio Code with the MCP extension/agent enabled

Step-by-step:

  1. Clone the repository: git clone https://github.com/davidlbl/catalog-api-mcp-server.git cd catalog-api-mcp-server

  2. Build the MCP server JAR (example using Maven): mvn -B -DskipTests package // The resulting JAR should appear at: target/catalog-api-mcp-server-0.0.1-SNAPSHOT.jar

  3. Ensure you have a Java installation and JAVA_HOME set to the JDK path. Example (bash): export JAVA_HOME=/path/to/jdk export PATH=$JAVA_HOME/bin:$PATH

  4. Configure the MCP in VS Code (settings.json) as shown in the README, pointing LOCAL_PATH to the project root and JAVA_HOME to your JDK.

  5. Start the MCP server through the VS Code interface by selecting the configured server and choosing start. The agent will load the configured tools and you can begin interacting with the local catalog API server.

Additional notes

Notes:

  • The MCP server is started by running the catalog-api-mcp-server-0.0.1-SNAPSHOT.jar JAR with the Java runtime specified by JAVA_HOME.
  • The LOCAL_PATH placeholder should reflect your local development directory where the catalog-api-mcp-server project is located.
  • If you modify the tool definitions or the build output path, update the args in the mcp_config accordingly.
  • Common issues include mismatched Java versions, incorrect path escaping for Windows vs. Unix-like systems, and ensuring the Maven/Gradle build completes successfully to generate the JAR.
  • The documentation link provided in the readme points to Spring AI tooling APIs for reference on available tool patterns.

Related MCP Servers

Sponsor this space

Reach thousands of developers