The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains standard instructional language for guiding agent behavior in a design review context. No malicious overrides, safety bypasses, or system prompt extraction attempts were found.- [INDIRECT_PROMPT_INJECTION]: Ingestion points: The skill is designed to evaluate external UI implementations and code provided in the user context. Boundary markers: No explicit delimiters or instructions to ignore embedded commands are defined for the analyzed content. Capability inventory: The skill has no access to dangerous tools such as shell execution, filesystem writes, or network operations. Sanitization: None. While the skill processes untrusted input, the lack of exploitable capabilities renders this surface harmless.- [DATA_EXFILTRATION]: No network operations (curl, wget, fetch) or access to sensitive local file paths (e.g., .ssh, .aws) were detected.- [REMOTE_CODE_EXECUTION]: The skill does not download or execute remote scripts. References to modern tech stacks (React 19, Tailwind) are architectural recommendations and do not involve runtime package installation.

frontend-design-review