The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external, untrusted data in the form of rendered web or iOS screens for visual analysis.
Ingestion points: Rendered components, screenshots, and local implementations provided by the user (SKILL.md).
Boundary markers: The instructions lack explicit delimiters or "ignore embedded instructions" warnings for the agent when processing user-provided visual content.
Capability inventory: The skill is restricted to analytical text output and does not utilize tools for subprocess execution, file system writing, or network exfiltration.
Sanitization: There are no instructions for sanitizing or filtering instructions that might be embedded within the UI content (e.g., text elements in a screenshot).
[SAFE]: The skill references external resources on uizze.com and github.com/uizze/uizze for its interactive tool, source code, and license. These are consistent with the skill's stated purpose and author information.

ui-slop-score