adobe-policy-guardrails

Pass

Audited by Gen Agent Trust Hub on Apr 8, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements security-focused features such as credential scanning patterns (detecting 'p8_' prefixes and JWTs) and content policy pre-screening for Adobe Firefly prompts.
  • [SAFE]: All provided code snippets are instructional or defensive in nature, focused on environment validation and resource management.
  • [SAFE]: Tool access is limited to standard operations required for the stated purpose of linting and CI configuration, with no evidence of malicious command execution or data exfiltration.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 8, 2026, 08:45 AM