The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructions include installing the kc-agent binary via a Homebrew tap (kubestellar/tap). This resource is provided by the project author and is a standard requirement for the skill's multi-cluster management functionality.
[PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection as it facilitates the processing of data from external Kubernetes environments.
Ingestion points: The kc-agent MCP server ingests Kubernetes resource data (e.g., Pods, Services, ConfigMaps) into the agent context (SKILL.md).
Boundary markers: No specific delimiters or "ignore" instructions for the ingested API data are defined within the skill file itself.
Capability inventory: The skill enables the agent to interact with Kubernetes APIs, which may include read or write capabilities depending on the user's active kubeconfig context (SKILL.md).
Sanitization: The skill does not specify any sanitization, validation, or filtering processes for the content retrieved from the Kubernetes API.

kubestellar-console