Operon OS Agenaut

Guides

MCP security: tool poisoning, prompt injection and how to defend an agent

An agent reads every tool's description before it acts. That makes the description an input the server controls. Most MCP attacks start there.

Updated 14 November 2026

The attacks

Why the model cannot be the defence

A model cannot reliably tell an instruction from data, because to a model both are text. Filters and system prompts reduce the rate; they do not make it zero. The defence has to be a control the text cannot talk its way past.

Defences that hold

  1. Do not show the agent the server's text. Give each tool your own name and description, and keep only the structure of its inputs.
  2. Classify tools by what they do, and decide by that class.
  3. Pin what you approved. Keep a fingerprint of each tool; if it changes, switch it off until someone looks.
  4. Hold actions that send data out, and scan their contents for things that look like secrets.
  5. Show the reviewer the exact request, with hidden characters made visible.
  6. Keep credentials out of the agent.
  7. Limit where the gateway will connect: public https only, no redirects.

How Operon OS Agenaut applies them

Agents connected through Agenaut read Agenaut's own tool names and descriptions. You can read the server's text, and adopt it if you choose, with hidden characters removed. Agenaut suggests a risk class and you confirm it. If a server changes a tool after you switched it on, the tool is switched off until you check. A call that would send out something that looks like a password, key or card number waits for you, and you are told what kind of thing was seen. Agenaut is a net for common cases, not a proof; the approval step is what catches the rest.

Try it. Operon OS Agenaut is in beta. Request access, or read the docs to see how an agent is connected.

Related guides

Terms used here are explained in the glossary.