The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting untrusted data from the Kubernetes cluster environment.
Ingestion points: The script scripts/remote_k8s.py retrieves cluster metadata (node status, pod counts, storage classes) using kubectl get -o json.
Capability inventory: The skill has infrastructure management capabilities, including executing shell commands via subprocess.run (in scripts/remote_k8s.py), ssh operations, and helm / make commands.
Boundary markers: The instructions do not define delimiters or specific 'ignore' instructions for the agent when handling cluster data.
Sanitization: The remote_k8s.py script performs structural validation by parsing JSON but does not sanitize content values extracted from the cluster.
[COMMAND_EXECUTION]: The skill executes system commands to interact with the Kubernetes cluster and manage the deployment process.
Evidence: scripts/remote_k8s.py uses subprocess.run to call kubectl and ssh. The SKILL.md file provides instructions for running make setup, make install, and make port-forward.
[REMOTE_CODE_EXECUTION]: The documentation includes commands that download and execute scripts from a remote source.
Evidence: The file references/topologies.md contains the command curl https://raw.githubusercontent.com/helm/helm/main/scripts/get-helm-3 | bash.
Context: This targets the official repository of a well-known service (Helm) for tool installation.

paidf-orchestration-setup