The Agent Skills Directory

[SAFE]: The skill consists of detailed instructions for an AI agent to generate React and Next.js frontend code. It establishes design constraints (e.g., typography rules, color palettes) and coding standards (e.g., use of Framer Motion, performance guardrails).
[COMMAND_EXECUTION]: The skill specifies that the agent should provide installation commands (e.g., npm install) for missing dependencies found during package.json verification. This is a standard and expected interaction for a developer-oriented coding assistant.
[EXTERNAL_DOWNLOADS]: The instructions reference reputable third-party libraries (e.g., Radix UI, Phosphor Icons) and a well-known image placeholder service (picsum.photos). These resources are used according to standard industry practices for web development.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to adapt its output based on user chat prompts, creating a potential surface for indirect instructions.
Ingestion points: User chat prompts (Section 1) are used to dynamically adjust baseline design parameters.
Boundary markers: The instructions do not define explicit delimiters or "ignore embedded instructions" warnings for the user-supplied data.
Capability inventory: The skill is authorized to output shell commands for package management based on project context.
Sanitization: No explicit sanitization or validation of user input is specified in the instruction set.

design-taste-frontend-v1