website-testimonials
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a process for the agent to handle untrusted data from customer quotes and testimonials.
- Ingestion points: External content is sourced from emails, surveys, support tickets, and review platforms (like G2 or Trustpilot) and transcribed into
src/data/testimonials.ts(referenced in SKILL.md). - Boundary markers: The instructions focus on verbatim transcription but do not specify the use of delimiters or instructions for the agent to ignore potential commands embedded within the testimonials.
- Capability inventory: The skill's defined capabilities are limited to data organization and UI rendering; it does not include dangerous operations such as arbitrary command execution, dynamic code evaluation, or unauthorized network exfiltration.
- Sanitization: The skill relies on the project's base layout to serialize data into JSON-LD schema, which typically involves standard character escaping and structural validation.
Audit Metadata