container-security-auditor

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [No Code] (SAFE): The file SKILL.md contains only metadata and descriptive text. No Python, JavaScript, or shell scripts are included.
  • [Command Execution] (SAFE): The skill requests Bash tool permissions, which are appropriate for a security auditing tool. No malicious commands are hardcoded in the skill.
  • [Indirect Prompt Injection] (SAFE): The skill identifies a potential ingestion point for untrusted container data. 1. Ingestion points: Reads container configuration files. 2. Boundary markers: Absent. 3. Capability inventory: Bash, Grep, Read, Write, Edit. 4. Sanitization: Relies on agent-side filtering.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:08 PM