stylized-rendering

Pass

Audited by Gen Agent Trust Hub on Jun 21, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill contains no shell commands, subprocess calls, or execution instructions. Code snippets are restricted to HLSL shader math and documentation.- [REMOTE_CODE_EXECUTION]: No remote script execution or unverified dependency downloads were found. External GitHub projects are mentioned neutrally by name for educational reference.- [DATA_EXFILTRATION]: No network operations, sensitive file path access, or credential exposure patterns were identified.- [PROMPT_INJECTION]: The content is purely technical and contains no instructions designed to override the agent's system prompt or bypass safety filters.- [CREDENTIALS_UNSAFE]: No hardcoded API keys, tokens, or private secrets were detected within the documentation or example shaders.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 21, 2026, 12:31 AM
Security Audit — agent-trust-hub — stylized-rendering