phishing-pretext
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data from web searches and public organizational pages to model corporate voice and phrasing.
- Ingestion points: The workflow specifically instructs the agent to search the web for official organizational pages, press releases, leadership pages, and policy documents (SKILL.md, Step 2 and 3).
- Boundary markers: The skill lacks explicit instructions to ignore or strip potential malicious instructions that might be embedded in the retrieved web content.
- Capability inventory: The skill is configured for web research and text generation. While it does not perform automated file-writes or network operations beyond search, the resulting lures are high-fidelity artifacts designed for social engineering.
- Sanitization: There is no documentation of input sanitization or validation of the content retrieved during the research phase.
Audit Metadata