cheatengine -bridge
Connect Cursor, Copilot & Claude directly to Cheat Engine via MCP. Automate reverse engineering, pointer scanning, and memory analysis using natural language.
claude mcp add --transport stdio miscusi-peek-cheatengine-mcp-bridge python MCP_Server/mcp_cheatengine.py
How to use
This MCP server provides a bridge between an AI agent and Cheat Engine via the MCP protocol. It exposes a suite of memory inspection, analysis, debugging, and structure-discovery tools designed to let the AI read memory, follow pointer chains, disassemble code, dissect structures, and set breakpoints through a standardized JSON-RPC interface over standard input/output. The server is intended to be paired with a client configuration that points to the local named pipe or through the MCP client loader, enabling you to ask questions like reading memory, disassembling functions, or discovering RTTI-based class names in the target process. The included toolset mirrors the Cheat Engine workflow, but is driven by AI prompts to accelerate memory analysis and game-reverse tasks.
How to install
Prerequisites:
- Python 3.10 or newer installed on Windows (as required by the Cheat Engine bridge). If you plan to run the bridge directly, Python and pip should already be configured in your PATH.
- Install Python dependencies
pip install -r MCP_Server/requirements.txt
If you want to install manually (without the requirements file):
pip install mcp pywin32
- Run the MCP server
# From project root, using the recommended path to the bridge script
python MCP_Server/mcp_cheatengine.py
Note: This server relies on Cheat Engine in DBVM mode and the Windows Named Pipes mechanism, so this setup is intended for Windows systems with Cheat Engine installed.
Additional notes
Tips and common considerations:
- Windows only: The bridge uses pywin32 to interface with Cheat Engine via Named Pipes. Ensure Cheat Engine is configured correctly and that the DBVM memory access settings do not trigger system protections.
- The MCP server is designed to be run as a local service; configure your MCP client to point to the server using the provided mcp_config.json example in your client.
- If you customize paths, ensure the Python script path is accessible and that the working directory contains the MCP_Server folder structure.
- When troubleshooting, verify connectivity with the ping tool and confirm the server reports the expected version (e.g., 11.4.x) and a successful connection state.
- If you modify or extend tools, keep the tool namespace consistent with the MCP_Bridge_Command_Reference documentation found under AI_Context.
Related MCP Servers
awesome-claude-skills
A curated list of awesome Claude Skills, resources, and tools for customizing Claude AI workflows
mcpcat-typescript-sdk
MCPcat is an analytics platform for MCP server owners 🐱.
IoT-Edge
MCP server for Industrial IoT, SCADA and PLC systems. Unifies MQTT sensors, Modbus devices and industrial equipment into a single AI-orchestrable API. Features real-time monitoring, alarms, time-series storage and actuator control.
work-memory
Never lose context again - persistent memory management system for AI-powered workflows across multiple tools
slack
Session-based Slack MCP for Claude and MCP clients: local-first workflows, secure-default HTTP.
cheatengine -bridge
🔗 Connect AI to Cheat Engine for fast memory analysis, enabling quick mods and audits without the tedious manual work.