AI Skill Security (Preview)
Learn how Prisma AIRS AI Skill Security identifies security and governance risks in
AI skills before they are used in your enterprise.
| Where Can I Use This? | What Do I Need? |
|---|
- Prisma AIRS (AI Skill Security)
|
- No License required for preview
|
Prisma AIRS AI Skill Security helps you identify security and governance risks in AI
skills before they are used within your enterprise. A skill is a reusable package that
extends the capabilities of an AI agent. Skills must include a
SKILL.md definition and can include supporting scripts,
prompts, configuration files, and other resources. Because skills may execute code,
access enterprise resources, or interact with external services, you should review them
before trusting them for use.
AI Skill Security statically analyzes each uploaded skill package, identifies potential
risks, evaluates those findings against your organizational policy, and produces an
Allowed or Blocked verdict.
AI Skill Security performs four primary steps during every scan:
- Upload the skill package.
- Analyze the contents of the skill and all referenced files.
- Evaluate findings against the configured security rules.
- Produce an Allowed or Blocked verdict along with detailed
findings.
Each completed scan includes:
- A fingerprint uniquely identifying the uploaded skill bundle
- Overall scan verdict
- Rule outcomes
- Vulnerabilities detected
- File locations associated with each finding
- Descriptions explaining the associated risk
