The Agent Skills Directory

[Prompt Injection] (SAFE): The skill contains technical instructions for testing and does not attempt to override AI safety constraints or extract system prompts.
[Data Exposure & Exfiltration] (SAFE): No hardcoded credentials, sensitive file access (e.g., .env or SSH keys), or suspicious network requests were identified. Use of databases is limited to local testing environments.
[Obfuscation] (SAFE): The code and documentation are clear and free of Base64, zero-width characters, or other hidden content techniques.
[Unverifiable Dependencies & Remote Code Execution] (SAFE): The skill references local development tools like phpunit and artisan. It does not download or execute scripts from remote or untrusted sources.
[Dynamic Execution] (SAFE): Mocking and faking patterns used are standard for Laravel testing and do not involve unsafe dynamic code generation or execution of untrusted input.

testing-laravel