The Agent Skills Directory

[SAFE]: The skill provides structured instructions for the AI to transform JSON input into ASCII TUI and Pencil MCP design specifications. It does not utilize any shell commands, file system operations, or network requests.
[INDIRECT_PROMPT_INJECTION]: The skill identifies a data ingestion surface through its Input Model (specifically the modelValue and state.error fields). However, because the skill lacks any high-privilege capabilities (no file writing, no network exfiltration, and no tool execution beyond formatting), the security risk is negligible. The rendering rules also include sanitization instructions such as truncating error strings to fit the output width.

tui-calendar