README.md•3.85 kB
# CS2 RCON MCP
[](https://www.python.org/downloads/)
[](https://cursor.sh)
[](https://opensource.org/licenses/MIT)
A Model Context Protocol server for CS2 RCON management.
## Description
This project provides a Model Context Protocol (MCP) server interface for managing CS2 game servers via RCON. It allows remote control and monitoring of CS2 servers through a standardized protocol.
## Features
- Manage your CS2 server in natural language
- RCON command execution
- Manage workshop maps (host, list, change) - [Explore Workshop Maps](https://steamcommunity.com/app/730/workshop/)
- SSE-based communication
- Docker support
## Available Tools
| Tool | Short Description |
|------|-------------------|
| `rcon` | Execute any RCON command |
| `status` | Get current server status |
| `list_workshop_maps` | List all workshop maps on the server |
| `host_workshop_map` | Host a workshop map by its ID |
| `workshop_changelevel` | Change the map to a given workshop map |
## Installation
### Environment Variables
- `HOST`: CS2 server IP
- `SERVER_PORT`: CS2 server port
- `RCON_PASSWORD`: RCON password
### Docker (recommended)
Pull the Docker image from GitHub Container Registry:
```bash
docker pull ghcr.io/v9rt3x/cs2-rcon-mcp:latest
```
### Docker Environment Variables
When running with Docker, you can set the environment variables in two ways:
1. **Directly in the command**:
```bash
docker run -p 8080:8080 \
-e HOST=your_server_ip \
-e SERVER_PORT=your_server_port \
-e RCON_PASSWORD=your_password \
ghcr.io/v9rt3x/cs2-rcon-mcp:latest
```
2. **Using a `.server-env` file**:
Create a file named `.server-env` with the following content:
```
HOST=your_server_ip
SERVER_PORT=your_server_port
RCON_PASSWORD=your_password
```
Then run the container like this:
```bash
docker run -p 8080:8080 --env-file .server-env ghcr.io/v9rt3x/cs2-rcon-mcp:latest
```
This provides users with an alternative method to set environment variables, making it easier to manage sensitive information like passwords.
### Connecting from Visual Studio Code (GitHub Copilot)
To configure Visual Studio Code to work with the MCP server, follow these steps:
1. **Start the MCP Server**: Ensure that your MCP server is running before attempting to connect from VS Code.
2. **Open Visual Studio Code**: Launch VS Code and ensure that you have the GitHub Copilot extension installed and configured.
3. **Configure GitHub Copilot**:
- Change the mode from "Ask" to "Agent" mode.
4. **Add MCP Server Configuration**:
- Click on the toolbox icon in the upper left corner of the Copilot prompt.
- Select "Add MCP Server" and choose the option for **HTTP - server-sent events**.
5. **Enter the Server URL**:
- For the URL, input: `http://localhost:8080/cs2server/sse`. This is the endpoint for the MCP server's SSE connection.
### Alternative: Connecting from Cursor (or any other MCP-Client)
1. Start the MCP server
2. Configure Cursor's MCP settings by creating or updating `~/.cursor/mcp.json`:
```json
{
"mcpServers": {
"cs2server": {
"url": "http://localhost:8080/cs2server/sse"
}
}
}
```
3. In Cursor, open the MCP panel (usually in the sidebar)
4. The server should automatically connect using the configured URL
Once connected, you can manage your server in natural language.
Example prompts:
1. "Add 5 bots to the server and start a competitive match on de_dust2"
2. "What's the current server status? How many players are connected and what map are we on?"
Happy fragging! 😊
