plannotator-visual-explainer
Generate self-contained HTML visualizations with Plannotator theming. Use for implementation plans, PR explainers, architecture diagrams, data tables, slide decks, and any visual explanation of technical concepts. Plans and PR explainers follow Plannotator's prescriptive approach; all other visual content delegates to nicobailon/visual-explainer.
Security Assessment
About plannotator-visual-explainer
Generates self-contained HTML visualizations themed with Plannotator tokens, covering implementation plans, PR explainers, architecture diagrams, data tables, slide decks, and other visual explanations of technical concepts. Content is routed down one of three paths based on type. Implementation plans, design docs, and proposals follow a prescriptive Plan path; PR walkthroughs and diff reviews follow a prescriptive PR path; everything else (diagrams, tables, decks, recaps, comparisons) is delegated to the nicobailon/visual-explainer skill with Plannotator theme tokens applied as a color and typography override.
Each path reads its own reference files before generating. The Plan path consults design-system.md and svg-patterns.md and builds documents from ordered sections such as a header, a three-to-five card summary strip, a phase-based timeline with no time estimates, inline SVG architecture or data-flow diagrams, HTML and CSS mockups, dark-theme code blocks for architecturally significant interfaces, a risks-and-mitigations table with severity badges, and open-question callouts. The PR path reads design-system.md and pr-components.md and assembles a header meta strip, a TL;DR card, a why and before-after comparison, collapsible per-file tour cards with NEW, MOD, or DEL badges, a tiered risk map, focus callouts, a test plan checklist, and an optional rollout section, using Pierre diffs via CDN for syntax-highlighted inline diffs.
Output is always delivered through Plannotator's annotation UI rather than a browser open command. Plans and proposals that a user should approve or deny are delivered with the gate flag (plannotator annotate <file> --gate), while informational content is delivered without it. A shared design philosophy governs all paths: whitespace is treated as a feature with generous padding, one idea is shown per viewport, relationships and sequencing are shown through diagrams and timelines rather than described, and time estimates are never attached to phases.
FAQ
What content types does this skill handle?
It produces visualizations for implementation plans, design docs, proposals, PR explainers, diff reviews, architecture diagrams, data tables, slide decks, project recaps, comparisons, and general visual explanations. Each type is routed to one of three paths.
How is the output delivered to the user?
Always through Plannotator's annotation UI, never with open or xdg-open. Plans and proposals are delivered with `plannotator annotate <file> --gate` so the user can approve or deny, and everything else is delivered with `plannotator annotate <file>` as informational content.
When does it delegate to the visual-explainer skill?
For everything other than plans and PR explainers, such as architecture diagrams, data tables, slide decks, recaps, and comparisons. It follows visual-explainer's structure and component classes, overriding only the color and typography layer with Plannotator tokens.
Does the skill include time estimates in plans?
No. Timelines show phases, sequence, and dependencies but never attach hour or day estimates. This is part of the design philosophy applied across all paths.
What reference files should be read before generating a plan?
For the Plan path, read references/design-system.md for theme tokens, typography, and component patterns, and references/svg-patterns.md for inline SVG building blocks like architecture diagrams, flowcharts, and data flow.
Install plannotator-visual-explainer
Quick Setup:
- Copy the skill folder to
.claude/skills/ - Claude will automatically detect and use the skill
Repository
backnotprop/plannotator