The Agent Skills Directory

[Overall Security Status] (SAFE): The content consists entirely of technical documentation, best practices, and code examples for the Laravel Filament PHP framework. No indicators of malicious intent or dangerous capabilities were identified.
[Prompt Injection] (SAFE): No instructions designed to override system prompts or bypass safety filters were detected. The language is purely instructional and framework-oriented.
[Data Exposure & Exfiltration] (SAFE): There are no hardcoded credentials, sensitive file paths, or network requests targeting non-whitelisted domains. The examples provided use generic service names (e.g., ExternalApiService).
[Obfuscation] (SAFE): All text and code snippets are human-readable. No Base64, zero-width characters, or homoglyph-based evasion techniques were found.
[Unverifiable Dependencies & RCE] (SAFE): While the skill mentions various Laravel/PHP ecosystem tools (Filament, Livewire, Alpine.js), it does not include commands to download and execute arbitrary scripts or install untrusted packages at runtime.
[Indirect Prompt Injection] (SAFE): The skill does not define any ingestion points for untrusted external data that would influence agent reasoning or execution flow. It serves as a static knowledge base for the agent.

filament-pro