WRG

You can deploy and use the Weapon Recoil Generator MCP Server to generate recoil patterns for different firearms and visualize them in real time. This server exposes two core tools through HTTP endpoints, letting you iterate on weapon balance efficiently during development.

How to use

You deploy the MCP server and call its endpoints to generate recoil data and visualize patterns. The two primary endpoints are related to weapon recoil generation and recoil visualization. Run the server locally, then access the endpoints to generate data for a machine gun, pistol, or shotgun, and plot the resulting recoil trajectories.

How to install

# Prerequisites: Python and pip
# 1) Clone the MCP server repository
git clone https://github.com/Hyeongseob91/mcp-server.git

# 2) Move into the project directory
cd mcp-server

# 3) Install dependencies
pip install -r requirements.txt

# 4) Start the server (development mode with automatic reload)
uvicorn main:app --reload --host 0.0.0.0 --port 8000

Additional sections

Configuration and usage with the local desktop automation tool involve adding an MCP server entry to the desktop client’s configuration. Here is a representative snippet you can adapt to your environment to point the client at the local server:

{
  "mcpServers": {
    "wrg": {
      "command": "python /path/to/mcp-server/main.py --http"
    }
  }
}

Key features you can leverage once the server is running include generating recoil data for different weapon types and visualizing recoil coordinates in a 2D plot. Use the following tool endpoints via your MCP client or HTTP client to access these capabilities.

• Weapon Recoil Generation (WRG) for machine guns, pistols, and shotguns produce recoil trajectory data based on the number of shots. Call the corresponding endpoints to obtain trajectory data for your balance tests.

• Recoil Visualizer (RVZ) renders recoil data as a 2D scatter plot to help you assess the pattern visually and iterate on tuning quickly.

• MCP Endpoints: /wrg/* for recoil data generation and /rvz/* for recoil visualization. These endpoints are designed to be easy to consume from backend, frontend, AI engineers, and game developers.

• Remote & Local Deployment: You can deploy the same API locally or to cloud environments that support MCP, enabling scalable testing and iteration.

• Debugging: Since the server communicates over HTTP, enable verbose logging and use the automatic reload feature during development. If you need deeper inspection, attach a debugger to the server process or review console errors.

Notes on configuration, security, and troubleshooting

• API authentication is planned for future use to control external access. Prepare to integrate an API key mechanism when it becomes available.

• For local testing, ensure the server is reachable at your chosen host and port (0.0.0.0:8000 is common for development). Check your firewall rules if you access the endpoints from another machine.

• If you are using a desktop client, restart the client after adding a new MCP server entry so that the /wrg and /rvz endpoints are recognized.

Available tools

weapon_recoil_generation

Generates recoil trajectory data for a specified weapon type and shot count, returning coordinates that describe the recoil path.

recoil_visualizer

Visualizes recoil trajectory data as a 2D scatter plot to facilitate quick inspection and iteration of recoil patterns.