competitor-analysis

Pass

Audited by Gen Agent Trust Hub on Jul 21, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists exclusively of instructional Markdown and interface configuration. It does not include any scripts, binary files, or system commands.
  • [PROMPT_INJECTION]: The skill was evaluated for indirect prompt injection surfaces. It instructs the agent to analyze external data sources such as competitor websites, pricing pages, and reviews. While these external sources could theoretically contain adversarial content, the skill itself provides no mechanisms to bypass safety filters or ignore previous instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 21, 2026, 05:32 AM
Security Audit — agent-trust-hub — competitor-analysis