learning-adaptive-testing

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [Prompt Injection] (SAFE): No instructions designed to bypass safety filters or override system prompts were detected in the metadata or documentation.
  • [Data Exposure & Exfiltration] (SAFE): The code does not perform any network requests or access sensitive local files. It only references a local 'framework' path for imports.
  • [Obfuscation] (SAFE): Content is clear, human-readable, and free from encoded strings, zero-width characters, or homoglyphs.
  • [Unverifiable Dependencies & Remote Code Execution] (SAFE): No external package installations or remote script executions (e.g., curl|bash) are present.
  • [Indirect Prompt Injection] (SAFE): The skill defines an 'input_data' parameter but does not currently process or interpolate this data into any prompts or executable contexts, presenting no active attack surface.
  • [Dynamic Execution] (SAFE): There is no use of eval(), exec(), or runtime compilation techniques.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:17 PM