pricing

Pass

Audited by Gen Agent Trust Hub on Oct 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The pricing page teardown functionality fetches and processes external URL content, creating an indirect prompt injection attack surface.
  • Ingestion points: External pricing pages (HTML/text) analyzed in references/pricing-page-teardown.md.
  • Boundary markers: Explicitly instructs the agent to treat fetched content as untrusted and to never follow instructions within it.
  • Capability inventory: No risky capabilities (file writing, shell access, network exfiltration) are present.
  • Sanitization: The skill relies on behavioral instructions to ignore malicious payload text within analyzed pages.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 11, 2026, 07:56 PM