The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external design system files based on configuration paths, which creates a potential surface for indirect prompt injection.
Ingestion points: The agent is instructed in SKILL.md to read the file located at the path defined by the toolbox.build_task.design_system_path configuration variable.
Boundary markers: The instructions do not include delimiters or specific commands to treat the content of the design system file as untrusted data or to ignore instructions embedded within it.
Capability inventory: The skill environment permits the use of powerful tools including Read, Grep, Glob, and Bash.
Sanitization: There is no evidence of validation, sanitization, or filtering performed on the content of the design system file before processing.

visual-validator