qa-split-testing-levels-pyramid

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill acts as a conceptual framework for QA testing strategy. It does not contain executable code, shell commands, or network operations. All instructions are focused on logical categorization of test scenarios.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data in the form of feature descriptions and risk scenarios.
  • Ingestion points: Feature descriptions and risk scenarios provided by the user or prior analysis phases.
  • Boundary markers: None explicitly defined in the prompt text.
  • Capability inventory: The skill orchestrates other analysis skills (e.g., scan-automation-project) but lacks direct high-privilege capabilities like arbitrary file writing or network exfiltration.
  • Sanitization: Not applicable as the skill performs logical categorization rather than data processing for execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 04:46 PM
Security Audit — agent-trust-hub — qa-split-testing-levels-pyramid