The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes user-provided interface descriptions and flows for evaluation. This ingestion of untrusted data introduces a risk of indirect prompt injection where malicious instructions could be embedded in the design content.
Ingestion points: User-supplied interface descriptions and screen flows defined in the SKILL.md documentation.
Boundary markers: No clear delimiters or sanitization steps are documented to isolate the input data from agent instructions.
Capability inventory: Evaluation against 168 principles and detection of UX antipatterns.
Sanitization: No explicit sanitization or input validation mechanisms are described for the untrusted data.
[EXTERNAL_DOWNLOADS]: The skill documentation provides an installation command, npx skills add uxuiprinciples/agent-skills, which downloads external functional logic from the vendor's repository.
[DATA_EXFILTRATION]: Mentions an optional API connection to uxuiprinciples.com to provide enriched findings and citations, which involves transmitting the user's interface context to the author's external service for processing.

uxui-principles