The Agent Skills Directory

[PROMPT_INJECTION]: The skill creates an attack surface for indirect prompt injection through its data ingestion process.
Ingestion points: The skill reads user feature descriptions from command arguments, scans the local codebase (CSS, JavaScript, and templates) using the Explore agent, and analyzes user-provided screenshots.
Boundary markers: The skill does not use specific delimiters or protective instructions when incorporating data from external files into its prompts or the generated HTML.
Capability inventory: The skill possesses the ability to create directories, copy and write files, and launch parallel sub-agents to perform visual design tasks.
Sanitization: There is no explicit sanitization or validation of the content retrieved from the codebase or screenshots before it is used to generate the wireframe prototype or provide context to sub-agents.
[EXTERNAL_DOWNLOADS]: The visual design phase allows for the inclusion of Google Fonts via CSS imports, which is a standard practice and utilizes a well-known service.
[COMMAND_EXECUTION]: The skill uses basic shell commands such as mkdir and cp for standard file and directory management tasks related to creating the wireframe output directory.

wireframe