perf-theory-tester

Pass

Audited by Gen Agent Trust Hub on Oct 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill provides standard procedural instructions for performance testing, focusing on baseline management and hypothesis validation. No instances of obfuscation, credential harvesting, or unauthorized network access were found.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it uses external input to drive significant system actions. 1. Ingestion points: Hypotheses provided as part of the experiment workflow. 2. Boundary markers: No specific delimiters are used to separate user-provided data from system instructions. 3. Capability inventory: The skill is designed to modify files and execute benchmark commands. 4. Sanitization: No sanitization or validation logic is defined for the changes or commands generated from the experimental data.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 9, 2026, 09:02 PM