The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill downloads and installs the pup CLI tool using Homebrew from a repository hosted by the vendor. This tool is required for interacting with APM services and verifying instrumentation configuration.
[COMMAND_EXECUTION]: Multiple shell commands are utilized to perform diagnostics, including:
kubectl for inspecting pod specifications and executing commands within containers.
pup for authentication and querying APM service status.
curl for generating test traffic to a locally forwarded port.
brew for managing the installation of diagnostic dependencies.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data to drive its logic, presenting a theoretical attack surface:
Ingestion points: The skill reads the output of kubectl get pod and pup apm services list to verify instrumentation status.
Boundary markers: No specific delimiters are used to separate the external command output from the agent's logic.
Capability inventory: The skill possesses the ability to execute shell commands (kubectl, pup, curl, brew) and perform network operations via curl and pup.
Sanitization: No explicit sanitization or validation of the ingested command output is performed before it is used to decide the next execution step.
[CREDENTIALS_UNSAFE]: The skill instructs the user to set a DD_APP_KEY environment variable as a fallback mechanism. The instructions explicitly warn against writing raw keys into chat or files, which follows secure credential management practices.

verify-ssi