Problem-Solving Techniques

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • SAFE (SAFE): The skill consists exclusively of informational Markdown documentation and does not contain any executable components or malicious instructions.
  • NO_CODE (SAFE): No code files, package manifests, or automated scripts are present in the skill package.
  • PROMPT_INJECTION (SAFE): The mental models provided (e.g., Inversion Exercise, Scale Game) are for user assistance and do not attempt to manipulate the agent's core instructions or safety filters.
  • EXTERNAL_DOWNLOADS (SAFE): The attribution to the 'microsoft/amplifier' repository refers to a trusted source and does not involve automated downloads.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:15 PM