Prism Scanner
Scans AI agent skills, plugins, and MCP servers for security vulnerabilities using 39 detection rules.
Security scanner purpose-built for AI agent ecosystems. Analyzes MCP servers, agent skills, and plugins across three layers: static analysis of tool definitions, behavioral analysis of runtime patterns, and ecosystem analysis of inter-tool interactions. Covers 39 detection rules spanning prompt injection, data exfiltration, privilege escalation, and resource abuse. Outputs structured vulnerability reports with severity ratings and remediation guidance.
Source
Repository: https://github.com/aidongise-cell/prism-scanner
Maintain Prism Scanner?
Let people know it's listed here — add the badge (live metrics, light/dark aware) or a plain link to your README or docs.
[Prism Scanner on getagentictools](https://getagentictools.com/mcp/aidongise-cell-prism-scanner?ref=badge)