cisco-ai-defensecisco-ai-defenseOfficialยท7 Agent Skills Included

skill-scanner

Scan AI agent skills for hidden malware and prompt injection

Scans AI agent skills for prompt injection, data exfiltration, backdoors, and malicious code before you install them. Combines pattern rules, YARA signatures, behavioral dataflow analysis, and LLM review to catch hidden threats while filtering out false alarms. Plugs into CI/CD pipelines, pre-commit hooks, and GitHub Actions so risky skills get blocked automatically.
npx skills add cisco-ai-defense/skill-scanner --all -g -y

All Skills in This Repository (7)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install Skill Scanner?โ–ผ

Run `npx skills add cisco-ai-defense/skill-scanner --all -g -y` in your terminal to install all tools in this suite globally.

How to check if an AI skill is safe?โ–ผ

Point the scanner at any skill folder or GitHub repo and it flags prompt injection, credential theft, and hidden malicious code with a severity report.

Can Skill Scanner run in CI/CD pipelines?โ–ผ

Yes. It outputs SARIF for GitHub Code Scanning, supports fail-on-severity exit codes, and includes a reusable GitHub Actions workflow.

Does Skill Scanner work with Claude Code and Cursor skills?โ–ผ

Yes. It supports OpenAI Codex and Cursor skill formats, and a lenient mode scans Claude Code commands and plain markdown skill repos.

Do I need an API key to use Skill Scanner?โ–ผ

No. Core static and behavioral scanning works offline; API keys are only needed for optional LLM analysis, VirusTotal, and Cisco AI Defense features.

Related Repositories in Software Engineering

View All in Software Engineeringโ†’