NYC

manimce-best-practices

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE] (SAFE): No security issues were identified during the analysis of the 31 files.
  • Prompt Injection: The skill does not contain any instructions aimed at bypassing AI safety protocols or overriding system prompts.
  • Data Exposure & Exfiltration: No hardcoded credentials, sensitive file paths, or suspicious network requests were found.
  • Unverifiable Dependencies: The skill correctly identifies the 'manim' package and standard scientific libraries (numpy, scipy) as dependencies.
  • Dynamic Execution: While the skill provides Python code for rendering animations, it does not employ unsafe dynamic execution of untrusted input or hidden payloads.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:41 PM