The Agent Skills Directory

[SAFE]: The skill is a functional development tool that follows industry best practices for UI/UX auditing. It does not contain any malicious patterns, obfuscation, or unauthorized data exfiltration logic. The guidance for installing dependencies via standard registries and configuring system libraries is transparent and appropriate for the intended use case.
[INDIRECT_PROMPT_INJECTION]: The skill operates by ingesting content from external websites and code repositories, which represents a potential attack surface for indirect prompt injection.
Ingestion points: External preview URLs (via browser automation tools), git diff outputs, and pull request descriptions as instructed in SKILL.md and references/browser-tools-reference.md.
Boundary markers: The instructions do not define specific delimiters or "ignore embedded instructions" warnings for the content retrieved from external sources.
Capability inventory: The skill has access to Bash, Read, Grep, and Glob tools, as well as browser automation for screenshots, console monitoring, and DOM interaction.
Sanitization: There is no explicit instruction to sanitize or validate the content retrieved from external sites before processing it.
Risk Assessment: Given the skill's purpose as an auditing tool, processing this external data is a functional requirement. The risk is minimized by the structured review methodology which keeps the agent focused on specific UI/UX criteria.

design-review