The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses curl (Linux) and curl.exe (Windows) to make authenticated requests to the Datadog API. These commands are necessary for the skill's primary function of monitoring and debugging.
[EXTERNAL_DOWNLOADS]: The skill fetches data from official Datadog API endpoints (e.g., api.datadoghq.com). These network operations are directed at a well-known service and are consistent with the skill's purpose.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external logs and monitors, which could potentially contain untrusted content.
Ingestion points: The skill reads output from the Datadog API via curl in SKILL.md and references/windows.md.
Boundary markers: There are no explicit markers defined to separate API output from agent instructions.
Capability inventory: The agent context includes shell command execution (curl, jq, ConvertFrom-Json).
Sanitization: No explicit sanitization or validation of the API response data is documented.

datadog