
Visual Explainer
Turns explanations, diffs, plans and data into self-contained HTML visual pages you can open in a browser.
- Repository
- nicobailon/visual-explainer
- Stars
- 10,741
- Licence
- MIT
- Files
- 15 files
What it does
Once installed, the agent builds a single HTML page that explains a topic mainly through diagrams, charts and labeled figures, with short captions instead of long text. It can make architecture diagrams, code-change or plan reviews, project recaps, comparison tables, slide decks and narrated MP4 videos. It renders the page locally and opens it in a browser, and readers can click parts of the page to ask questions or leave comments that get sent back to the agent.
Great for
- Developers who want to explain architecture, a pull request, or a refactor to teammates visually
- Product or project leads who need a quick visual recap or plan review
- Anyone who prefers seeing a system drawn out rather than reading a long document
Skip it if
- Tasks that need plain text or a Markdown-only output, since the default output is HTML with figures (the skill only writes Markdown when asked)
- Sensitive material you do not want rendered into a local file, because the page content and any reply data are written to disk and may be opened in a browser
Try asking
- “Draw the architecture of this service as a visual explainer with the request path and cache layer”
- “Make a visual review of the diff on my current branch with risks and next steps”
- “Turn this project's recent commits into a visual recap page”
- “Create a comparison page of these two database options with trade-offs”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 15 original files, the MIT licence and an attribution note.
visual-explainer.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip visual-explainer.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip visual-explainer.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Name the audience and goal in your request so the agent picks the right page shape and depth
- Ask for a specific output (diagram, slide deck, video) to avoid the default page when you need something else
- Review the generated page in your browser and use the comment or ask features to request targeted changes
What's inside
- .claude-plugin/plugin.json
- commands/compare.md
- commands/diff-review.md
- commands/fact-check.md
- commands/generate-slides.md
- commands/generate-video.md
- commands/generate-visual-plan.md
- commands/generate-web-diagram.md
- commands/plan-review.md
- commands/project-recap.md
- commands/state-machine.md
- extension.ts
- kit/ask.css
- kit/ask.js
- SKILL.md
Redistributed unmodified from nicobailon/visual-explainer under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.