audit-ux-and-save-files
Pass
Audited by Gen Agent Trust Hub on Jul 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns detected. The skill performs UX auditing tasks using browser automation in a restricted and structured manner.
- [COMMAND_EXECUTION]: Uses standard shell utilities (curl, find, grep, wc, ls, git) for environment checks and file management. These operations are scoped to the skill's internal workflow and project directories and do not involve arbitrary user-supplied input execution.
- [EXTERNAL_DOWNLOADS]: The skill is installed via standard package managers from the author's own repository (yigitkonur/skills-by-yigitkonur), which is consistent with the provided vendor context.
- [PROMPT_INJECTION]: The skill processes data from external applications via browser automation. The subagent prompt establishes clear boundaries and instructions to role-play as a persona, reducing the likelihood of obeying instructions embedded in the audited content. Ingestion occurs via the browser agent reading app content, with boundaries maintained by structured instructions in the subagent template.
Audit Metadata