social-proof-injection
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a system that ingests external data (prospect profiles, needs, and testimonials) and interpolates them directly into generated messages.
- Ingestion points: Prospect attributes like
stated_needs,pain_points, andobjections(referenced in theselect_social_prooffunction) and social proofcontent(referenced informat_testimonial). - Capability inventory: The skill facilitates the creation and formatting of sales outreach messages for bot responses.
- Boundary markers: None identified in the prompt templates or code snippets; content is directly interpolated into message strings.
- Sanitization: No escaping, validation, or filtering is applied to the prospect data or testimonial content before it is formatted for output in the
format_testimonialandformat_case_study_teaserfunctions.
Audit Metadata