The Agent Skills Directory

[SAFE]: The skill provides standard instructional content for a frontend developer persona, focusing on component ownership, state transitions, and accessibility validation.
[INDIRECT_PROMPT_INJECTION]: The skill processes external task data and has workspace write permissions, which represents a potential surface for indirect prompt injection. Ingestion points include task flow mapping and component boundaries defined in SKILL.md. Boundary markers are not present to isolate instructions from task data. Capability inventory includes workspace write access (sandbox_mode: workspace-write). Sanitization of external input is not specified. This attack surface is determined to be safe given the skill's primary function as a development tool.

frontend-developer