The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses Node.js to execute local scripts and manages browser processes for video rendering. It includes logic to locate system-installed browsers (Chrome/Edge) to perform encoding.
[EXTERNAL_DOWNLOADS]: The workflow depends on the playwright-core library for browser automation and requires the presence of a web browser for H.264 video encoding via WebCodecs.
[SAFE]: The implementation demonstrates strong security awareness through several mechanisms:
Input Validation: Detailed validation functions (validateCoverage, validateBrand, validateClips) verify the structure and content of CSV and JSON inputs.
Sanitization: All user-supplied text is escaped using escapeHtml before being rendered into the HTML dashboard, and scriptJson is used for safe data embedding.
File Integrity: SVG assets are scanned for malicious patterns such as scripts or remote URLs to prevent cross-site scripting (XSS) or data leaks.
Content Security Policy: The generated HTML reports use a restrictive CSP (Content Security Policy) that prevents unauthorized script execution and network requests.
Minimal Permissions: The renderer is designed to work offline without making network requests.

coverage-reel