competitor-offer-analysis

Pass

Audited by Gen Agent Trust Hub on Jun 29, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection because it is designed to ingest and analyze content from external, untrusted competitor URLs.
  • Ingestion points: The competitors input field in SKILL.md accepts a list of URLs that the agent is expected to visit and analyze.
  • Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions to prevent the agent from obeying commands found on external websites.
  • Capability inventory: The agent reads external web content and processes it to generate positioning recommendations.
  • Sanitization: There is no mention of sanitizing or escaping the content retrieved from external URLs before it is processed by the model.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 29, 2026, 04:36 PM
Security Audit — agent-trust-hub — competitor-offer-analysis