1 item
See what AI agents see. Detect prompt injection, risky tools, hidden instructions, and sensitive actions. GuardM8 is an AI-agent security firewall. Its single purpose is to inspect webpages locally for prompt injection, deceptive machine-readable content, risky WebMCP or browser-agent tools, sensitive-data fields, and consequential agent-triggered actions, then warn the user or request approval when appropriate.
rating_count is the Chrome Web Store ratings count, not a written-review count.