The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute various aws and kubectl commands to discover cluster state, resources, and configurations. These commands are necessary for the skill's primary purpose of migration assessment and are generally scoped to read-only operations (list, describe, get).
[INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it ingests untrusted data from the live EKS cluster (e.g., Ingress annotations and ConfigMaps).
Ingestion points: Data is ingested via kubectl get commands for Ingress, IngressClass, Deployments, and ConfigMaps across various namespaces.
Boundary markers: The skill does not explicitly define boundary markers or 'ignore' instructions for the data retrieved from the cluster during its analysis phase.
Capability inventory: The skill possesses the capability to execute shell commands (aws, kubectl), write to the local filesystem (~/ingress_migration/), and run a local Python script.
Sanitization: The provided tools/report_to_html.py script performs HTML escaping on the markdown content to mitigate risks when generating the visual report.
[EXTERNAL_DOWNLOADS]: The skill contains references to official documentation and GitHub repositories for AWS and Kubernetes (e.g., kubernetes-sigs.github.io, cert-manager.io). These are well-known, trusted sources and are used for informational purposes or manifest templates.

eks-ingress-migration