The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes a local Node.js utility script (scripts/lint.mjs) and instructs the agent to execute it to validate the design compliance of generated files. The script performs static analysis by reading files using node:fs and checking for regex patterns related to design tokens, contrast, and typography.
[EXTERNAL_DOWNLOADS]: The HTML templates (app-shell.html, landing.html) reference external font resources from Google Fonts (fonts.googleapis.com and fonts.gstatic.com). These are well-known, trusted services used for legitimate asset delivery.
[INDIRECT_PROMPT_INJECTION]: The skill processes project files through its linting script, creating a potential surface for indirect prompt injection if those files contained malicious instructions targeting the agent.
Ingestion points: Files passed as arguments to node scripts/lint.mjs in the project directory.
Boundary markers: None present in the script's output.
Capability inventory: The script uses node:fs for local file reading and outputs findings to the console.
Sanitization: The script performs regex-based scanning for design violations and does not execute the content of the files it analyzes.

perplexity-design