The Agent Skills Directory

[SAFE]: The skill serves a legitimate audit purpose, scanning Google Ads accounts for creative compliance issues such as broken URLs, ad disapprovals, and inappropriate content. The logic is implemented in clear-text Python scripts that utilize official APIs.
[EXTERNAL_DOWNLOADS]: The installation instructions in the README utilize curl to fetch the skill's components from the author's public GitHub repository (github.com/fourteenwm). As these are vendor-owned resources for the installation of the tool itself, this is considered normal and safe behavior.
[COMMAND_EXECUTION]: The skill executes a local Python script (ads_checker_audit.py) to perform the audit. The script is self-contained, uses well-known libraries (such as google-ads and gspread), and does not attempt to execute arbitrary commands or download additional executables at runtime.
[DATA_EXFILTRATION]: While the tool transmits audit data to Google Sheets, this is its primary documented function. The transmission is authorized via the user's own API credentials, and the analysis confirms no data is sent to unknown or unauthorized third-party servers.
[CREDENTIALS_UNSAFE]: The skill demonstrates safe credential handling by instructing the user to store sensitive API keys and OAuth tokens in an external configuration file (google-ads.yaml) rather than hardcoding them within the skill's instructions or scripts.

ads-checker