MCP-Dandan security monitor

MCP-Dandan is a security monitor for your AI assistant. It watches what your AI is doing behind the scenes and warns you when something looks risky. It is handy for anyone who lets an AI read files, run commands, or connect to other apps and wants a safety net.

What this MCP server does

You run MCP-Dandan on your own computer. It watches the messages your AI sends to its tools, like when the AI asks to read a file or run a command. It checks each request against several detection engines that look for command injection, file exposure, personal data leaks, data exfiltration, and tool misuse. If it finds a threat, it shows you in a desktop app and can block the action. You stay in control the whole time.

Click to zoom

What you can do with it

• See every tool call your AI makes in a live dashboard
• Get alerts when the AI tries to read sensitive files
• Block suspicious actions before they happen
• Detect when personal information might be leaked
• Spot unusual data transfers
• Catch when a tool is used in a way it was not meant for
• Add your own rules for what counts as personal data

Try asking your AI

• “Show me the latest tool calls from my AI”
• “Did anything suspicious happen in the last hour?”
• “Block that file read request”
• “Add a rule to flag any mention of social security numbers”

What it gives back to you

You get a live feed of tool calls and alerts in the desktop app. Each alert tells you what happened, which engine caught it, and how serious it is. You can then choose to allow or block the action. The app also keeps logs you can review later.

Before you start

Who it's for

Anyone who uses an AI assistant that can access files or run commands and wants to keep an eye on what it is doing.