The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's primary workflow involves reading and analyzing project source files (CSS, HTML, JSX, TSX). This creates a surface for indirect prompt injection if these files contain malicious instructions disguised as comments or data.
Ingestion points: Project files processed during review, polish, or build modes.
Boundary markers: The skill does not implement specific delimiters or instructions to disregard potential prompts embedded within the code being analyzed.
Capability inventory: The agent has the ability to read and write files and execute the included ui-check.mjs script.
Sanitization: There is no mechanism to sanitize or validate the content of processed files for prompt injection threats.
[COMMAND_EXECUTION]: The skill utilizes a local Node.js script, scripts/ui-check.mjs, to perform linting and static analysis tasks. While this script is a legitimate part of the skill's functionality, it represents an executable capability within the skill's environment.

khe-ui