The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external websites via Playwright and interprets image generation prompts from course metadata.
Ingestion points: The skill fetches content from user-provided URLs in scripts/scrape-tools.mjs and processes prompt specifications from the illustrations array in course-data.js.
Boundary markers: No explicit delimiters or instructions to ignore embedded content were found in the provided code snippets.
Capability inventory: The skill utilizes file system writes (fs.writeFile), network access (for scraping and QR code generation), and command-line execution (node, powershell).
Sanitization: No sanitization or validation logic is described for the content scraped from the web or the prompt metadata interpolated into the generation manifest.
[COMMAND_EXECUTION]: The skill provides templates for executing shell commands, including launching Chrome with remote debugging and a custom user data directory (--remote-debugging-port=9222, --user-data-dir), as well as running Node.js scripts for scraping and verification.
[EXTERNAL_DOWNLOADS]: The skill references external Node.js dependencies playwright and qrcode for its primary functionalities.

web-visual-assets