The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the installation of the anousss007/blatui package via Composer from a third-party source.
Evidence: composer require anousss007/blatui in the installation section.
[COMMAND_EXECUTION]: The skill executes multiple shell commands, including php artisan commands that trigger logic from the newly installed third-party package.
Evidence: php artisan blatui:init, php artisan blatui:add, and php artisan vendor:publish --tag=blatui-foundations.
[INDIRECT_PROMPT_INJECTION]: The skill analyzes the project environment and automatically writes files to the project directory, which creates a vulnerability surface if the agent processes untrusted project files.
Ingestion points: blatui:init scans project packages, CSS, and JS configurations (SKILL.md).
Boundary markers: None present; the skill assumes the project environment is trustworthy.
Capability inventory: Writing Blade files to resources/views/components/ui/, modifying resources/css/app.css and resources/js/app.js (SKILL.md).
Sanitization: No mention of validation or escaping for project-specific configurations before generating component code.

blatui-laravel-blade-components