The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The instructions describe fetching a reference values.yaml file from a public repository via curl at https://raw.githubusercontent.com/bjw-s-labs/helm-charts/main/charts/library/common/values.yaml to assist the agent in identifying configuration schemas.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection or input manipulation surface by incorporating external input directly into file generation and command execution contexts.
Ingestion points: External user input enters the context through the $ARGUMENTS variable to define the application name.
Boundary markers: Absent; no specific delimiters or shielding instructions are provided to enclose the input.
Capability inventory: The skill possesses file Write capabilities and Bash command execution privileges (e.g., kustomize build, make lint).
Sanitization: Absent; the guidelines do not instruct the agent to sanitize or validate the $ARGUMENTS variable against path traversal sequences or command execution characters before utilizing it to create directories or execute commands.

scaffold-app-template