perplexity

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • General (SAFE): No malicious patterns found. The skill documentation describes a search tool that uses an external API key provided via environment variables.
  • Indirect Prompt Injection (SAFE): The skill provides a surface for data ingestion (web search results), which is the intended behavior. No exploitable capabilities or malicious interpolation patterns were found in the analyzed documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 01:51 PM