The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection because it instructs the agent to navigate to, interact with, and deeply analyze untrusted external websites provided by the user.
Ingestion points: The reference URL provided by the user in the 'User Interview' step.
Boundary markers: None present. The instructions do not include delimiters or warnings to ignore embedded instructions in the analyzed page content.
Capability inventory: The skill utilizes file-read (to understand current project structure), file-write (to implement the design in the local codebase), and network/browser interaction (via Playwright MCP to navigate and click around external sites).
Sanitization: No sanitization or validation of the external content is described.
[COMMAND_EXECUTION]: The skill utilizes the Playwright Model Context Protocol (MCP) to perform complex interactions with live websites, including navigation, hovering, and clicking. While intended for design deconstruction, these capabilities could be exploited if the agent is successfully injected with malicious instructions from a visited page.
[EXTERNAL_DOWNLOADS]: The documentation references an external third-party tool repository: github.com/executeautomation/mcp-playwright. Users are expected to have or install this tool to enable the skill's functionality.

aesthetic-analysis