The Agent Skills Directory

[COMMAND_EXECUTION]: The scripts scripts/agent.py and scripts/process.py use the subprocess.run function to execute kubectl commands. This is the primary mechanism the skill uses to query the Kubernetes API for information about namespaces, pods, and network policies for security analysis.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted metadata from the execution environment, creating a vulnerability surface for indirect prompt injection.
Ingestion points: Data including resource names, labels, and selector configurations are ingested from the Kubernetes API via kubectl in scripts/agent.py and scripts/process.py.
Boundary markers: No explicit delimiters or boundary markers are used in the generated audit reports to separate cluster-sourced metadata from core agent instructions.
Capability inventory: The skill has the capability to interact with the cluster API through kubectl and perform file system writes to save audit reports.
Sanitization: There is no evidence of sanitization or validation of the content of resource metadata before it is processed by the scripts or included in the final report.

implementing-network-policies-for-kubernetes