The Agent Skills Directory

[SAFE]: No malicious code, obfuscated content, or persistence mechanisms were detected. The skill's behavior is entirely consistent with its described purpose of managing observability integrations.
[CREDENTIALS_UNSAFE]: The skill avoids hardcoding secrets, instead using clear placeholders like [datadog_api_key] and [datadog_app_key] for sensitive values, which is the correct security practice for documentation.
[EXTERNAL_DOWNLOADS]: Network operations are directed towards well-known observability services (e.g., Datadog) for the purpose of metric and event ingestion, which is part of the intended integration functionality.
[COMMAND_EXECUTION]: The skill utilizes SN-Execute-Background-Script and Bash for legitimate administrative tasks, such as auditing alert normalization rules and analyzing CI binding accuracy. These scripts are restricted to reporting and do not involve unsafe execution of untrusted input.

observability-integration