The Agent Skills Directory

[SAFE]: The skill provides legitimate documentation and boilerplate for modern web development. No malicious behavior, obfuscation, or sensitive data access was identified.
[PROMPT_INJECTION]: The skill includes operational instructions for AI agents in the 'BEFORE YOU START' section. These instructions are intended to ensure adherence to the provided architectural patterns and do not attempt to bypass safety guardrails or ethics protocols.
[EXTERNAL_DOWNLOADS]: The documentation references official installation commands for established tools and libraries like npm, bun, and the shadcn/ui CLI. These are standard dependencies for the target development environment.
[INDIRECT_PROMPT_INJECTION]: The skill presents an attack surface for indirect injection via template processing, though the risk is negligible.
Ingestion points: Documentation and templates in SKILL.md and the templates/ directory.
Boundary markers: Uses markdown formatting and code blocks for structure; no specialized boundary markers are used for template content.
Capability inventory: The agent is instructed to perform file system writes and package installations.
Sanitization: The content is static boilerplate; sanitization of dynamic user input is deferred to the agent's core safety logic.

tailwind-v4-shadcn