Get the FREE Ultimate OpenClaw Setup Guide →

OrbisPrismMCP

Orbis Prism is a small tool for Hytale modding assistance. It decomposes the official server (HytaleServer.jar), isolates its core logic, and provides an AI-assisted Intelligent Query Interface (MCP) for developers.

Installation
Run this command in your terminal to add the MCP server to Claude Code.
Run in terminal:
Command
claude mcp add --transport stdio orbisfactory-orbisprismmcp python -m orbis_prism_mcp \
  --env JAVA_HOME="Path to a Java 17-25 JRE/JDK (required for Hytale server compatibility and decompilation tools)" \
  --env PRISM_HOME="Default directory for global storage, typically ~/.prism"

How to use

Orbis Prism MCP exposes an AI-ready MCP server that integrates with the Prism toolkit and supports connecting AI agents such as Cursor or Claude. After installing Orbis Prism MCP, initialize your workspace to detect your Hytale installation, decompile server logic, and index the API and assets. Start the MCP server using the Prism CLI with prism mcp, which launches the MCP endpoint configured by the package. The server exposes a structured API surface (via MCP) that AI agents can query for server methods, assets, and decompiled class information, enabling natural language or scripted interactions with the game logic.

With the included tooling, you can switch decompiler engines (JADX by default or Vineflower), perform fast API searches with SQLite FTS5, and explore assets directly from the Assets.zip. As an AI-ready endpoint, the MCP server is designed to be integrated with agents like Cursor or Claude, enabling reasoning over decompiled server behavior, asset data, and API surfaces. To use it, install orbis_prism_mcp, initialize your workspace using prism ctx init --assets, and start the MCP server with prism mcp. Then connect your AI agent to the server’s MCP interface to begin querying game logic and assets.

How to install

Prerequisites:

  • Official Hytale installation (Launcher and game files)
  • Python 3.11+
  • Java 17-25 (required for server compatibility and JADX/Vineflower decompilers)

Install from PyPI:

pip install orbis_prism_mcp

(Optional for local development):

pip install -e .

Initialize and run:

prism ctx init --assets
prism mcp

If you prefer to run directly via Python module (advanced):

python -m orbis_prism_mcp

Note: Ensure JAVA_HOME and a compatible Java runtime are available in your environment, as the MCP server relies on decompilers that require Java.

Additional notes

Storage: Orbis Prism stores configuration and databases under ~/.prism by default. You can override the workspace with prism ctx init in a specific folder or by using the -w flag. Decompiler choice is configurable, with JADX as the default and Vineflower as an alternative. The MCP server is designed to be integrated with AI agents; ensure your agent can reach the running MCP endpoint. If you encounter issues connecting to your Hytale installation, verify that the installation path is accessible and that your Java runtime is correctly configured (JAVA_HOME).

Environment variables to consider:

  • PRISM_HOME: path to Prism storage (default ~/.prism)
  • JAVA_HOME: path to a Java 17-25 JRE/JDK
  • Optional: ASSETS_DIR if you need to override asset locations

Common issues:

  • Missing Java runtime: install a compatible JRE/JDK and set JAVA_HOME.
  • Unable to detect Hytale installation: ensure the official game files are present and accessible.
  • Decompiler failures: switch between JADX and Vineflower in Prism settings if you run into parsing errors.

Related MCP Servers

Sponsor this space

Reach thousands of developers