sketch

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or data exfiltration attempts were detected in the skill instructions or reference files.
  • [CREDENTIALS_UNSAFE]: The skill demonstrates safe secret management practices by explicitly forbidding the hardcoding of API keys and instructing users to use environment variables (os.environ["GEMINI_API_KEY"]) and .gitignore for .env files.
  • [EXTERNAL_DOWNLOADS]: The skill references several well-known and official libraries, including the google-genai SDK and image processing tools like Real-ESRGAN, imagehash, and Pillow. These are standard dependencies for the stated purpose of the skill.
  • [COMMAND_EXECUTION]: While the skill provides Python code snippets for the user to execute, it does not perform any autonomous command execution on the host system. The generated code is intended for the user's local development environment.
  • [INDIRECT_PROMPT_INJECTION]: The skill includes a robust 'Content Policy Guardrails' reference that outlines a five-layer policy stack to prevent the generation of harmful content, demonstrating high awareness of adversarial prompt risks.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 06:56 AM
Security Audit — agent-trust-hub — sketch