The Agent Skills Directory

[SAFE]: The skill enforces a strict security posture by managing API credentials through environment variables rather than command-line arguments, preventing secret exposure in process lists or shell history.
[SAFE]: Sensitive data protection is maintained by setting restricted file permissions (0o600) on the local token cache and all downloaded reports and logs, ensuring only the current user can access this data.
[SAFE]: The implementation includes robust redaction logic that automatically scrubs API keys, bearer tokens, and credentials embedded in URLs from request previews, error messages, and saved metadata.
[SAFE]: A safety-first approach is applied to resource mutations (Create, Update, Delete); the agent is instructed to perform a GET request and show a preview of the changes, requiring explicit user confirmation before executing the mutation with the --execute flag.
[SAFE]: Network operations are strictly scoped to the official Moloco Ads API domain (api.moloco.cloud). The tool is designed to avoid credential leakage by ensuring that authentication headers are never forwarded when downloading files from pre-signed storage URLs.
[SAFE]: The script uses only Python standard libraries and avoids dynamic execution patterns like eval() or exec(), significantly reducing the risk of code injection.

moloco-ads