The Agent Skills Directory

[SAFE]: The skill is designed for static analysis and auditing of Laravel codebases. It focuses on identifying 'smells' and security risks without performing dangerous operations or exfiltrating data.
[CREDENTIALS_UNSAFE]: The skill searches for sensitive files like .env to detect leaked credentials. This is its intended primary purpose as an auditor, and it includes robust safeguards such as redacting values in reports and advising on credential rotation.
[PROMPT_INJECTION]: The skill processes untrusted code from external repositories, creating a potential surface for indirect prompt injection. This risk is mitigated by the skill's instructions to use a structured finding format and require concrete evidence for all claims.
Ingestion points: The skill reads application controllers, configuration files, and environment files (.env) in the repo.laravel-app repository.
Boundary markers: The skill mandates a structured report format with specific ID patterns (AUD-LARAVEL-{nnn}) to organize its output.
Capability inventory: The skill utilizes file system reading and pattern matching (grep) capabilities.
Sanitization: The skill explicitly requires redacting secret values and emphasizes providing remediation guidance instead of auto-applying changes.

laravel