MCP server · Security
BlastRadius MCP server
by Jimil-Joshi
Check how dangerous an AI action is before it runs, hide secrets, and keep a tamper-proof record.
BlastRadius is a safety helper that sits between your AI assistant and the tools it can use. Before your AI runs something risky, like deleting files or dropping a database table, this helper scores how bad it could be and can stop it. It is handy for anyone who lets an AI touch real files, databases, or cloud accounts.
What is an MCP server? The 30-second version
On its own, your AI can only chat with you. An MCP server is a small helper program that gives your AI a new skill or a connection to something else. This particular helper adds a safety layer: it watches what your AI is about to do, checks it against rules, and hides secrets before they leak. You keep using your AI the same way, and BlastRadius quietly checks each action in the background.
What this MCP server does
When your AI wants to run a command or use a tool, it asks BlastRadius first. BlastRadius reads the request, works out what the command would really do once the shell has rewritten it, and gives it a danger score from 0 to 100. If the score is low, the action can go ahead. If it is high, BlastRadius blocks it or asks for a signed approval from a human. Along the way it also spots passwords, keys, and personal details in the text and replaces them with a redacted marker.
What you can do with it
- Score a command before it runs and see how dangerous it is
- Block destructive actions like deleting a whole disk or dropping a database
- Hide API keys, passwords, and personal details from text
- Ask for a signed human approval before high-risk actions
- Stop an approval from being reused twice
- Check the audit log for signs of tampering
- See live security numbers and the active policy
Try asking your AI
- “Is it safe to run rm -rf /var/lib/postgresql on this machine?”
- “Check this text for secrets before I send it: Bearer AKIAIOSFODNN7EXAMPLE”
- “Do I need approval to run terraform destroy in production?”
- “Show me the current security posture and any recent blocked actions”
What it gives back to you
You get back a clear answer in the chat: a danger score, a severity label like SAFE or CRITICAL, and a short list of reasons. For secret checks you see what was found and a cleaned-up version with the sensitive parts replaced. For policy checks you get a decision of ALLOW, BLOCK, or REQUIRE_CONFIRMATION, plus the rule that fired. Audit checks tell you whether the record is intact or has been changed.
Before you start
What you need
- Node.js 22 or newer installed on your computer
- An AI assistant that supports MCP servers, like Claude Desktop or Cursor
Good to know
It can block or stop actions your AI wants to take, so a blocked command will not run until you approve it or change the policy.
Install it with your AI
Add BlastRadius MCP server to your AI, no technical skills needed
You don't install anything by hand. You copy one prompt, paste it into an AI that can work on your computer, and it checks, installs and connects the server for you, asking you when it needs something.
Sign in to get the install prompt
Members get a ready-made prompt that lets the Claude desktop app check BlastRadius MCP server, install it and connect it for them, step by step. You don't need any technical skills: you copy, paste and answer a few questions. Your connected AI can also find and install any of the 4,210 MCP servers here for you.
Who it's for
Developers, IT staff, and anyone who lets an AI assistant run commands or touch real systems and wants a safety net.
