angle-mort

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The instructions specify that the engine should "dare a lot" and that "audience safety guards" (garde-fou d'audience) may restrict the required creative divergence. This is a stylistic persona guideline for the skill's primary purpose and does not represent an attempt to bypass the model's underlying safety filters.
  • [NO_CODE]: The skill is entirely descriptive and instruction-based, containing only markdown documentation without any associated scripts, binaries, or software library dependencies.
  • [SAFE]: The skill provides links to the author's YouTube channel and Substack newsletter. These are well-known services provided for standard promotional purposes and do not involve obfuscated or malicious URLs.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 09:21 PM
Security Audit — agent-trust-hub — angle-mort