The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to perform multiple shell operations, including directory creation (mkdir -p), file copying (cp -r), and executing a Python script located in the user's home directory (~/.claude/skills/ui-ux-pro-max/scripts/search.py) to obtain design system recommendations.
[EXTERNAL_DOWNLOADS]: The project setup involves npm install, which retrieves packages from the official npm registry. Additionally, the use of Puppeteer in export scripts triggers the download of a browser binary from a well-known source.
[DYNAMIC_EXECUTION]: The core functionality involves generating React component files (.jsx) based on user input and then executing them via a development server (npm run dev) and a build process (npm run build:standalone).
[INDIRECT_PROMPT_INJECTION]: The skill features a 'Research Checkpoint' that performs web searches to gather data and statistics. This data is incorporated into the presentation context and slides, which represents a surface for indirect prompt injection if the ingested web content contains adversarial instructions.
Ingestion points: Web search results (Step 2.5) and user-provided notes (Step 2).
Boundary markers: None explicitly defined in the provided templates for data ingestion.
Capability inventory: The skill can execute shell commands, install packages, and interact with the browser via agent-browser.
Sanitization: There is no explicit sanitization step described for the data retrieved during the research phase before it is used for code generation.

slides-generator