micronaut-fun
MCP Server and Web application to provide Search engine for Micronaut Documentation and Guides
claude mcp add --transport stdio micronaut-projects-micronaut-fun docker run -i micronautfun:latest \ --env OPENSEARCH_HOST="http://localhost:9200" \ --env MICRONAUT_FUN_PORT="8080" \ --env OPENSEARCH_PASSWORD="" \ --env OPENSEARCH_USERNAME=""
How to use
Micronaut Fun exposes an MCP server endpoint at /mcp that allows you to query and manipulate Micronaut documentation resources and guides through the MCP protocol. The server is designed to enable semantic search over Micronaut docs and guides, and to provide tooling prompts and context to help you extract relevant information during conversations. You can leverage the included tools to search Micronaut documentation or guides, generate summaries, and retrieve specific sections or topics. The setup described in the repository uses OpenSearch as the backing store, and the web application is built with Micronaut and Java. To interact with the MCP server, point your MCP client at the /mcp endpoint and use the available actions to fetch documentation resources, run searches, or request structured contexts for your prompts.
How to install
Prerequisites:
- Java JDK 11 or newer (for building/running Micronaut app) or a prebuilt binary/image if you prefer Docker.
- Docker (for running the OpenSearch container and/or the Micronaut Fun server container).
- Optional: OpenSearch running locally or via Docker if not using the included container.
Install options:
-
Local development (Gradle-based app, not containerized in this setup):
- Ensure JDK is installed.
- Clone the repository.
- Run the app locally:
- ./gradlew build
- ./gradlew run
- Ensure OpenSearch is available at http://localhost:9200 (or adjust config).
-
Docker-based deployment (recommended if you want an isolated environment):
- Ensure Docker is installed.
- Start OpenSearch (example provided by the repo): docker run -d --name micronautfunopensearch -p 9200:9200 -p 9600:9600 -e "discovery.type=single-node" -e "plugins.security.disabled=true" -e "OPENSEARCH_INITIAL_ADMIN_PASSWORD=DummyPassword#1233" opensearchproject/opensearch:2.19.3
- Build or pull the Micronaut Fun container image (as named in your deployment): docker pull micronautfun:latest
- Run the Micronaut Fun container (as shown in the example): docker run -d --name micronautfun -p 8080:8080 -e OPENSEARCH_HOST=http://localhost:9200 micronautfun:latest
-
If you prefer a local Gradle run without Docker:
- Ensure OpenSearch is running.
- Run: ./gradlew run
Note: The exact image name and environment variables may vary based on your deployment setup. The README indicates running OpenSearch via Docker and removing a specific feature flag in the application properties for a full run.
Additional notes
Tips and common issues:
- Ensure OpenSearch is accessible at the host/port your config expects (default http://localhost:9200).
- If you modify application-dev.properties to enable the Micronaut app with OpenSearch, remove the flag micronaut.opensearch.enabled=false as indicated in the docs.
- When running via Docker, expose the correct ports for both the app and OpenSearch (as shown in the docker examples).
- The MCP endpoint is at /mcp; use an MCP client to send requests to this path and interpret the structured responses for context-building tasks.
- If you encounter connection errors to OpenSearch, verify network connectivity between containers or adjust the OPENSEARCH_HOST to the correct address (e.g., a docker network alias).
- For development, you can mock OpenSearch responses or run a local OpenSearch instance to speed up iterations.
- If you switch to a non-Docker setup, ensure the environment variables (like OPENSEARCH_HOST) are set in your environment or application properties accordingly.
Related MCP Servers
mcp-for-beginners
This open-source curriculum introduces the fundamentals of Model Context Protocol (MCP) through real-world, cross-language examples in .NET, Java, TypeScript, JavaScript, Rust and Python. Designed for developers, it focuses on practical techniques for building modular, scalable, and secure AI workflows from session setup to service orchestration.
tda
TDA - Thread Dump Analyzer (for Java). Analyze your Thread Dumps with a GUI or use it as MCP Server.
sonarqube
SonarQube MCP Server
wanaku
Wanaku MCP Router
WigAI
Bitwig Controller Extension that provides an MCP Server for AI Agent control
SchemaCrawler-AI
Free database schema discovery and comprehension tool