Picrew

Build and refine claim-led academic or technical slide decks with Academic Deck Compiler. Use when the user asks for tasteful research slides, PPTX, Beamer, paper talks, technical portfolios, or multi-agent slide iteration.

Build HTML-first researcher or project profile slides that borrow the information architecture of polished academic homepages such as PRISM, with dense but clean identity, publication, project, and news-style layouts.

Use the vendored beautiful-html-templates library as a tone-first reference or HTML template source for tasteful academic, technical, profile, and product decks. Trigger when the user asks to use beautiful-html-templates, wants more varied HTML slide aesthetics, or needs a template shortlist before rendering.

Review rendered contact sheets for HTML-first academic or technical decks, deciding what to keep, discard, and change next based on aesthetics, readability, evidence clarity, and slide-to-slide cadence.

Run a disciplined design-context workflow for HTML-first slides: verify facts, collect assets, choose visual grammar, make 2-slide showcases, compare directions, and critique outputs to avoid AI-slop.

Choose, crop, and judge images for tasteful research slides, prioritizing proof surfaces over decorative screenshots, portraits, logos, or generic AI visuals.

Select, crop, and annotate screenshots, paper figures, project pages, and result images so evidence slides prove claims with tasteful large proof surfaces.

Translate a slide DESIGN.md or high-end web reference into renderer-ready visual grammar tokens for HTML-first academic or technical decks.

Prompt set for Daily Blog scoring, synthesis, highlights, and conservative tone guard. Supports multilingual templates via language-suffixed files.

Use when the user has already provided source material and needs a cited, evidence-constrained answer from that material. Not for external retrieval.

Use when the user wants a fast architecture map from a GitHub repository URL or local code folder, and specifically needs a direct HTML visualization of module-to-module relationships. This skill is lightweight: the model inspects the repo, extracts modules/relations, and writes a self-contained HTML report (no external fetch required) plus a small JSON and summary file.