customer-journey
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill operates according to its stated purpose without any malicious commands, obfuscation, or sensitive data access. The external source link matches the author's verified personal branding.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources during the research phase.
- Ingestion points: Web search results and user-provided websites or keywords.
- Boundary markers: No explicit delimiters are specified for separating external content from instructions.
- Capability inventory: Restricted to web search and text generation; no evidence of subprocess execution, file writing, or network exfiltration tools.
- Sanitization: The instructions do not define specific sanitization or filtering for retrieved web content.
Audit Metadata