blockquote
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from user-supplied URLs and provides a prompt to automatically apply fixes based on that data.
- Ingestion points: The
urlparameter in thestart_scantool and the content retrieved from that URL (SKILL.md). - Boundary markers: The skill documentation does not mention the use of delimiters or 'ignore embedded instructions' warnings for the reports generated from external content.
- Capability inventory: The skill includes a specific prompt,
fix_top_issues, which directs the agent to 'Apply the top fixes to the project.' This capability allows data from an external, potentially untrusted source to influence local project changes. - Sanitization: There is no evidence of sanitization or intent validation for the content retrieved from external URLs before it is presented to the agent for action.
Audit Metadata