The Agent Skills Directory

[SAFE]: The skill follows best practices for data-driven auditing. It uses a restricted query language (GAQL) to fetch advertising metrics and focuses on generating human-readable reports and workspace documentation. No suspicious network activities, credential harvesting, or obfuscation techniques were detected.- [SAFE]: All file system interactions are limited to the skill's defined workspace (workspace/ads/) for the purpose of maintaining context and generating audit deliverables. The logic is transparent and aligned with the stated purpose of performing an ads audit.- [SAFE]: The acquisition of Google Ads data is performed through legitimate API queries meant for analysis. There is no evidence of command injection or the use of dynamic execution environments that could be exploited by remote data.

google-ads-audit