The Agent Skills Directory

[SAFE]: The skill implements a standard and legitimate interface for querying Datadog observability data through its public API, fulfilling its stated purpose.
[CREDENTIALS_SAFE]: Authentication is correctly handled via environment variables (DD_API_KEY, DD_APP_KEY). The documentation directs users on how to obtain and set these keys without including any hardcoded secrets.
[DATA_EXFILTRATION]: Network communication is targeted at Datadog's official domains. The site region is configurable via DD_SITE (e.g., datadoghq.com, datadoghq.eu), which is standard practice for Datadog integrations and does not constitute a data exfiltration risk in this context.
[COMMAND_EXECUTION]: The script processes command-line arguments to parameterize API requests. It does not utilize dangerous functions like eval() or exec() to process user input, and no shell injection vulnerabilities were found.
[REMOTE_CODE_EXECUTION]: No patterns for downloading or executing remote code were detected. Dependencies are standard and managed through manifest files.

datadog