skills/wojons/skills/security-scan/Gen Agent Trust Hub

security-scan

Pass

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/analyze-security.sh performs unsafe string concatenation when generating its JSON scan plan. It interpolates unsanitized command-line arguments (e.g., --target, --container, --iac) directly into the critical_actions array. If these generated command strings are subsequently executed by an agent or a user, it could lead to arbitrary shell command injection or parameter injection attacks.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection (Category 8) because it processes external identifiers and contexts without sanitization.
  • Ingestion points: The CLI arguments --target, --container, --iac, and --context in scripts/analyze-security.sh allow external data to enter the agent's context.
  • Boundary markers: None. User-controlled variables are inserted directly into the JSON heredoc without delimiters or "ignore instructions" warnings.
  • Capability inventory: The skill uses command -v to check for the presence of multiple system tools (npm, snyk, trivy, zap, etc.) and generates actionable shell commands intended for security scanning.
  • Sanitization: No input validation, escaping, or schema enforcement is applied to the arguments before they are included in the generated output.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 28, 2026, 10:30 AM