html-effectiveness-editor
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines patterns for creating interactive tools such as triage boards and prompt tuners using standard HTML/JS/CSS snippets.
- [DATA_EXFILTRATION]: The skill uses the Clipboard API (
navigator.clipboard.writeText) as a primary channel to return work results from the generated UI to the agent. This is a transparent and user-controlled process for data transfer. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided data (e.g., ticket lists, JSON samples) within the generated UI templates. This represents a standard surface for data processing without evidence of malicious intent or bypassing agent constraints.
Audit Metadata