The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local tools such as node for a screenshot script (ds-shot.mjs) and the glab CLI for merge request management. These are standard operations for a development automation tool.
[PROMPT_INJECTION]: The skill identifies as a surface for indirect prompt injection due to its ingestion of external Figma design data and variables. Ingestion points: Figma design context and variable definitions. Boundary markers: None present in the instructions. Capability inventory: Shell command execution via node and glab, and file system write access. Sanitization: No explicit data sanitization steps are documented.
[SAFE]: Human-in-the-loop checkpoints are established for plan approval, style verification, and final acceptance, mitigating the risks of automated execution.

restore