context-adaptation

Pass

Audited by Gen Agent Trust Hub on Apr 18, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is entirely descriptive and instruction-based, focusing on the concept of 'Context Adaptation' from the 'Huangdi Neijing'. It does not invoke any tools, perform network requests, or access the file system.
  • [PROMPT_INJECTION]: Instructions are restricted to the domain of environmental adaptation and do not attempt to override the AI's core behavior or safety protocols.
  • [DATA_EXFILTRATION]: There are no patterns suggesting the collection or transmission of sensitive data.
  • [EXTERNAL_DOWNLOADS]: The skill does not reference or download any external packages or scripts.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 18, 2026, 02:24 PM