patch-prioritization

Pass

Audited by Gen Agent Trust Hub on Jul 20, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns or security risks were identified. The skill provides standard process guidelines for security engineering tasks.
  • [EXTERNAL_DOWNLOADS]: The skill retrieves vulnerability intelligence from well-known services, including the FIRST.org EPSS API and the CISA KEV catalog. These references are used solely for information gathering and data enrichment.
  • [PROMPT_INJECTION]: The skill includes a dedicated safety section instructing the agent to disregard instructions embedded in untrusted vulnerability data. 1. Ingestion points: Vulnerability inventory findings and ticket descriptions (SKILL.md). 2. Boundary markers: Absent. 3. Capability inventory: File inspection using Read, Grep, and Glob tools (SKILL.md). 4. Sanitization: Absent, but mitigated by explicit instructions to prioritize framework logic over external content.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 20, 2026, 02:11 PM
Security Audit — agent-trust-hub — patch-prioritization