Build Stac DSL screens and themes from product requirements with safe defaults and reusable templates. Use when users ask to create or refactor StacScreen files, map UI requirements to Stac widgets/actions/styles, or scaffold new screen/theme files.
Create, update, or plan a Stencila theme for documents or published sites. Use when asked to choose a theme direction, write or patch theme.css, recommend semantic or module token families, customize site navigation or branding, tune PDF and print page tokens, align web, Python, and R plots with a Stencila design system, list available builtin tokens with `stencila themes tokens`, or validate a theme file with `stencila themes validate`.
Compose, polish, generate, and QA publication-grade CCF paper figures, tables, captions, palettes, panel maps, Python plotting code, and manuscript visual layout integration from supplied data/results. Use for figure/table layout, visual QA, palette selection, LaTeX figure/table placement, multi-panel design, creative data visualization, data-analysis plots, pie/donut charts, bar charts, volcano plots, correlation heatmaps, composite dashboards, source-data traceability, and making visuals fit naturally in the paper. Do not design experiments, invent results, write manuscript prose as the main task, or perform final submission compliance.
Build and edit small, personal visual tools and artifacts — dashboards, trackers, calculators, data visualizations, charts, simple landing pages, and slide decks the user wants for THEMSELVES. This is the right skill whenever the user asks to "visualize this," "make a chart," or "build an artifact" for their own use, or to edit an app they already built here. Do NOT reach for a ui_show dynamic_page to fake an artifact — build a real persistent app here. NOT for complex, multi-user, or shippable products — those go to a real project folder with a coding agent (see Scope below).
Customize the app's visual theme — author design-token overrides (accent, background, surfaces, text, message-bubble colors) in the workspace ui/theme.json, validated by the assistant runtime and applied live to connected clients. Covers the token slots, the all-or-none override groups, the contrast floor, and reading back validation issues.
Agent Loop for pixel-perfect UI adjustments: extract a structured spec from the user's intent, apply changes one property at a time, self-verify via screenshot, and converge instead of ping-ponging. Designed for high-fidelity UI cloning and spacing/style iteration.
Create publication-quality scientific figures and plots using Python (matplotlib, seaborn, plotly). Supports bar charts, scatter plots, heatmaps, box plots, violin plots, survival curves, network graphs, and more. Use when user asks to plot data, create figures, make charts, visualize results, or generate publication-ready graphics. Triggers on "plot", "chart", "figure", "graph", "visualize", "heatmap", "scatter plot", "bar chart", "histogram".
Use when working with UI editor tools — sprite slicing, atlas tools, safe area setup, BaseView template creation, and UI utility components in F8Framework.
Shape non-trivial work before writing it — decide the modules, the seams, and the public API surface up front. Use when starting a feature, adding an integration, changing a public barrel file, crossing a package or native-interop boundary, or deciding where a seam should go. Produces a design sketch to approve before implementation, then hands off to code-guidelines and test-guidelines.