What it does
This MCP server scans AI agent skills for 25 attack classes and adds runtime monitoring. The publisher's description cites more than 1,316 findings.
It suits anyone who wants to check AI agent skills against these attack classes and monitor them while they run.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
- Runs code or changes an assistant's behaviour
Sources
- Official MCP Registry first seen 13 Nov 2026 · last seen 13 Nov 2026 · its page on the source
com.arcself/arc-security
Scan AI agent skills for 25 attack classes + runtime monitoring. 1,316+ findings.
v0.5.1As shown on registry.modelcontextprotocol.io when we last read it