defining-icp
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists entirely of instructional markdown content and references to well-known business resources. It does not include any scripts, command execution logic, or unauthorized network activity.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze user-supplied data about their business and customers, which is a common surface for indirect prompt injection.
- Ingestion points: User descriptions of their business segment and customer characteristics as requested in
SKILL.md. - Boundary markers: None identified. The instructions do not specify the use of delimiters or protective phrasing to isolate user input.
- Capability inventory: No high-risk capabilities such as file system writes, system command execution (subprocess/exec), or external network requests were found in
SKILL.md,references/artifacts.md, orreferences/guest-insights.md. - Sanitization: There are no instructions provided for the validation, filtering, or sanitization of external content.
Audit Metadata