The Agent Skills Directory

[SAFE]: The skill analyzes local project files to provide a report on testing quality. It does not perform network operations, exfiltrate data, or execute untrusted code.
[DATA_EXFILTRATION]: The skill scans configuration files such as .env and .env.testing. This is done specifically to verify database isolation practices (e.g., checking for SQLite in-memory configuration) which is a standard part of a Laravel testing audit. No sensitive data is extracted or transmitted externally.
[PROMPT_INJECTION]: The skill ingests untrusted data by scanning PHP source files in the project. This represents an indirect prompt injection surface.
Ingestion points: app/, routes/, and tests/ directories, and composer.json file.
Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded prompts in the analyzed code.
Capability inventory: The skill's capabilities are limited to scanning files and generating a structured audit report; it does not write files, make network requests, or execute commands.
Sanitization: No explicit sanitization of the ingested code is performed before analysis.

laraveldaily-testing-audit