The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes user-provided HTML reference files to extract design systems. This introduces a surface for indirect prompt injection where a malicious HTML file could attempt to influence the agent's behavior during the extraction process.
Ingestion points: External HTML files provided by the user and referenced via the $ARGUMENTS parameter in references/extract-design-system.md.
Boundary markers: The prompt structure uses backticks and headers to delimit the reference content but does not include explicit instructions to ignore potentially malicious commands embedded within the reference source code.
Capability inventory: The agent has capabilities for file system operations (read/write), version control execution (git), and deployment (vercel).
Sanitization: No specific sanitization or validation of the input HTML content is described.
[COMMAND_EXECUTION]: The skill instructs the agent to use standard development tools as part of the intended web design and deployment workflow.
Evidence: Mentions of git add . && git commit for tracking design iterations and vercel for final site deployment in SKILL.md and references/modernize-site.md.
[EXTERNAL_DOWNLOADS]: The skill references the use of well-known JavaScript and CSS libraries via CDNs to implement animations and interactive effects.
Evidence: References to GSAP, ScrollTrigger, Locomotive Scroll, and AOS in SKILL.md and references/build-landing-page.md.

nv-design