The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes helm, curl, and kubectl to perform administrative tasks within a Kubernetes environment and interact with the CAST AI cloud service. These commands are necessary for managing cluster resources and autoscaling policies.
[EXTERNAL_DOWNLOADS]: Fetches and installs the Workload Autoscaler components from the official CAST AI Helm repository. As CAST AI is a well-known service, these downloads are considered safe within the intended use-case.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process data from external sources, including API responses and Kubernetes events, to generate cost-saving recommendations and monitoring evidence. While this creates an attack surface for indirect prompt injection, it is considered low risk given the service type.
Ingestion points: API responses from api.cast.ai and event logs retrieved via kubectl (referenced in SKILL.md).
Boundary markers: None identified in the prompt instructions to delimit or instruct the agent to ignore potentially malicious embedded content in external data.
Capability inventory: The skill possesses file writing/editing capabilities and bash command execution (helm, kubectl, curl).
Sanitization: No explicit sanitization, validation, or filtering of the ingested external data is specified before the agent processes it to perform actions or generate output.

castai-core-workflow-b