The Agent Skills Directory

[SAFE]: The skill restricts all network communication to official Unity Ads domains (services.api.unity.com and monetization.api.unity.com) using HTTPS, preventing unauthorized data exfiltration to third-party hosts.- [SAFE]: A custom redirect handler (SafeRedirectHandler) is implemented to ensure that sensitive Authorization headers are automatically stripped if an API request is redirected to a different origin, such as a cloud storage provider.- [SAFE]: Secret management follows industry best practices by requiring owner-only file permissions (0400 or 0600) and providing redaction logic to prevent credentials from appearing in error logs or agent outputs.- [SAFE]: Every mutation (POST, PUT, PATCH, DELETE) requires a multi-step safety workflow including a pre-flight preview of the changes, explicit user confirmation, and a post-flight verification GET request to ensure the remote state matches the requested state.- [SAFE]: The underlying Python script includes rigorous path validation to prevent directory traversal or the injection of absolute URLs into API calls.

unity-ads