The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill implements a 'Research' phase where it navigates to external web pages using Chrome MCP tools to extract design language, layout geometry, and component styles.
Ingestion points: External data enters the agent's context through read_page, find, and javascript_tool calls in Phase 1 (SKILL.md).
Boundary markers: There are no instructions provided to wrap the ingested content in delimiters or to apply specific safety warnings to the agent regarding the external content.
Capability inventory: The skill possesses the capability to write local files (briefs and HTML) and execute shell commands to start a local server.
Sanitization: The skill lacks explicit instructions for sanitizing or filtering the content extracted from target websites before processing it into animation plans.
[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to manage a local preview environment in Phase 4 (SKILL.md).
Evidence: Execution of python3 -m http.server 8765 --directory <output-directory> & to serve generated files and lsof -i :8765 to check port availability.
[EXTERNAL_DOWNLOADS]: The skill generates HTML files that reference resources from a well-known external service.
Evidence: The generated animations include an @import statement to fetch typography from Google Fonts' official service.

css-animation