The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external project files, design tokens, and assets which serve as potential attack surfaces for instructions hidden in data.
Ingestion points: Instructions in SKILL.md, references/redesign.md, and references/review.md direct the agent to read target implementation files, existing project guidance, and design tokens.
Boundary markers: The instructions do not specify the use of delimiters or 'ignore' commands to isolate external content from the agent's core instructions.
Capability inventory: The skill has the authority to read and modify files (in tune and redesign modes) and control browser or app-control tools to inspect and interact with rendered interfaces.
Sanitization: There is no mention of filtering or sanitizing the content of the source code or design documentation before it is processed by the model.

ui-design