clone-ui
Pass
Audited by Gen Agent Trust Hub on May 14, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill includes defensive instructions to protect against prompt injection in third-party content. It explicitly warns the agent to ignore strings like 'ignore previous instructions' and 'exfiltrate environment' when they appear in source data, treating them as data rather than instructions.
- [EXTERNAL_DOWNLOADS]: The skill fetches assets such as images, icons, and favicons from user-specified target URLs to facilitate accurate UI cloning. It also recommends well-known tools like chrome-devtools-mcp to enhance fidelity, while ensuring the user remains in control of the installation process.
- [COMMAND_EXECUTION]: Local helper scripts (save-tool-result.ps1 and save-tool-result.py) are used to extract and save JSON data from tool outputs. These scripts are bundled with the skill and perform limited file operations within the workspace, with no network or unauthorized access capabilities.
- [DATA_EXFILTRATION]: The skill processes external website data. It contains a dedicated security section instructing the agent not to clone authenticated or private surfaces by default to prevent accidental exposure of sensitive user sessions or personal information.
Audit Metadata