The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill retrieves design guidelines from a public repository hosted by Vercel Labs. This operation fetches a markdown file containing the rules used for the auditing process.- [PROMPT_INJECTION]: The skill ingests and analyzes user-provided UI code (such as HTML, JSX, and CSS), which creates a potential surface for indirect prompt injection where instructions embedded in the audited data could affect agent behavior.
Ingestion points: User-provided files and glob patterns are read by scripts/design-guideline-checker.py and scripts/html-accessibility-scanner.py.
Boundary markers: There are no explicit markers or instructions defined in the skill to treat the audited content as untrusted data or to ignore instructions contained within it.
Capability inventory: The skill has the capability to read files from the local filesystem based on user input and report content snippets back to the agent.
Sanitization: The provided scripts perform regex-based pattern matching but do not filter or sanitize the file content before it is processed by the agent.

web-design-guidelines