skills

Create reveal.js HTML presentations as single-file artifacts. Use this skill whenever the user asks to create a presentation, slide deck, or slides using reveal.js — or when they mention "reveal.js", "revealjs", or want an HTML-based presentation (as opposed to PowerPoint/PPTX). Also trigger when the user wants to convert content (outlines, markdown notes, topics) into a reveal.js deck, or asks to edit/improve an existing reveal.js presentation. This skill covers the full reveal.js feature set: slides, vertical slides, fragments, auto-animate, code highlighting, markdown slides, math (KaTeX/MathJax), backgrounds (color, gradient, image, video, iframe), themes, transitions, speaker notes, layout helpers, lightbox, scroll view, PDF export, the React wrapper (@revealjs/react), and plugin creation. If the user mentions any of these features in the context of presentations, use this skill.

Review a pre-implementation contract or clarification summary as a feedback harness before coding starts. Use when a task already has a clarification artifact and you want a second-pass review of scope boundaries, repository grounding, validation quality, risk coverage, and approval edges before implementation proceeds.

Run a final pre-implementation go or no-go gate after clarification and contract review. Use when a task already has a clarification artifact and, optionally, a contract review verdict, and you want one explicit execution decision: go, go-with-caveats, or no-go before any code changes begin.

Build a pre-implementation harness for ambiguous or risky coding tasks by grounding the request in the repository, producing a structured impact map, surfacing ambiguities and risks, defining scope boundaries, and creating a validation-ready implementation contract before any code changes are made. Use when a task is broad, underspecified, cross-cutting, or likely to drift without an explicit planning checkpoint.

Run a structured four-phase engineering workflow (discovery, implementation, verification, release-readiness) with Human in the Loop checkpoints and explicit go/no-go gates. Use this whenever the user asks for full codebase review before changes, wants QUESTIONS.md-driven implementation, requests validation against approved decisions, or needs a final release gate verdict (Ready, Ready with caveats, Not ready).