The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on the pup CLI to perform core functions such as listing services, analyzing dependencies, and aggregating trace data. These commands are executed within the user's environment to interact with the Datadog API.
[EXTERNAL_DOWNLOADS]: The skill provides instructions for installing the pup utility via Homebrew (brew install pup) from a repository associated with the service provider. This is a standard installation procedure for a well-known monitoring tool.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection attack surface (Category 8) because it ingests and processes external monitoring data.
Ingestion points: External data enters the agent's context through the output of commands like pup traces search, pup traces aggregate, and pup apm services list (found in SKILL.md).
Boundary markers: No boundary markers or "ignore instructions" delimiters are used to wrap the output of the monitoring commands to prevent the agent from obeying instructions that might be embedded in service names or trace metadata.
Capability inventory: The agent can execute shell commands via the pup CLI and associated environment tools as described in SKILL.md.
Sanitization: There is no evident sanitization or filtering of the telemetry data (e.g., resource names, error messages) before it is analyzed by the agent.

dd-apm