author-info
Pass
Audited by Gen Agent Trust Hub on Aug 11, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill instructs the agent to ingest and analyze external data (rendered HTML and HTTP responses), creating a surface for indirect prompt injection.\n
- Ingestion points: Rendered HTML and HTTP response content (SKILL.md).\n
- Boundary markers: The instructions lack specific delimiters or directions to ignore instructions embedded within the analyzed content.\n
- Capability inventory: The skill is limited to auditing and flagging; it does not include scripts or commands for shell execution, file system modification, or network exfiltration.\n
- Sanitization: No sanitization or validation of the external content is described.
Audit Metadata