markdown-viewer-skills — Claude Code skill
markdown-viewer-skills — a router skill for diagrams written as Markdown text: PlantUML for UML and networks, Vega-Lite and ECharts for charts, HTML/CSS for infographics and cards. Useful when docs need figures.
MIT 202 installs ★ 35 updated 2026-10-21
npx skills add podo/design-agent-skills --skill markdown-viewer-skills -a claude-codeWhat markdown-viewer-skills does
The skill is a router. It reads the goal behind a request (service reliability, product metrics, software design, cloud architecture, planning and roadmaps, internal documents) and points to the matching engine reference. PlantUML handles UML, BPMN, ArchiMate plus network and security layouts with an icon stencil set; Vega, Vega-Lite and ECharts cover charts; infographic templates and bare HTML/CSS cover cards and system layouts. A library of checked examples ships with it. Upstream names the skill documd-visuals, and that is the name you invoke once it is in place.
Figures stay as text inside the Markdown and a docu.md-compatible viewer draws them on the spot, with nothing pre-rendered. Any other Markdown renderer shows the figure source as a code block, so without that viewer there is little to see.
Export to html, epub, docx or pdf is a separate path: Node 18+ and Chrome/Chromium, plus the npx @markdown-viewer/documd CLI, which the install does not fetch. Slides and math notation are out. Mermaid, draw.io and Graphviz are marked as not recommended; for sketch style the package sends you to excalidraw-agents365 and hand-drawn-diagrams.
What it helps with
- draw a UML class diagram inside a README
- map a cloud or network architecture as text
- chart product metrics with Vega-Lite
- lay out an infographic card in HTML/CSS
- describe a BPMN process without exporting an image
Example prompts for the agent
- Draw a UML class diagram for the payments module in
ARCHITECTURE.md - Build a Vega-Lite chart of monthly revenue and drop it into
metrics.md - Make a PlantUML network diagram for the staging setup
How to install markdown-viewer-skills
1Skills CLI (Vercel)
npx skills add podo/design-agent-skills --skill markdown-viewer-skills -a claude-code2Manually: put the files in place
~/.claude/skills/markdown-viewer-skills/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/markdown-viewer-skills/
1Skills CLI (Vercel)
npx skills add podo/design-agent-skills --skill markdown-viewer-skills -a codex2Manually: put the files in place
.agents/skills/markdown-viewer-skills/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add podo/design-agent-skills --skill markdown-viewer-skills -a cursor2Manually: put the files in place
.cursor/skills/markdown-viewer-skills/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install podo/design-agent-skills skills/markdown-viewer-skills2Manually: put the files in place
.github/skills/markdown-viewer-skills/SKILL.md1Skills CLI (Vercel)
npx skills add podo/design-agent-skills --skill markdown-viewer-skills -a gemini-cli2Platform CLI
gemini skills install https://github.com/podo/design-agent-skills --path skills/markdown-viewer-skillsManually: put the files in place
.devin/skills/markdown-viewer-skills/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add podo/design-agent-skills --skill markdown-viewer-skills -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about markdown-viewer-skills
How do I install markdown-viewer-skills in Claude Code?
Run npx skills add podo/design-agent-skills --skill markdown-viewer-skills -a claude-code. The skill lands at ~/.claude/skills/markdown-viewer-skills/SKILL.md.
What does markdown-viewer-skills do?
It routes a request to the right engine: PlantUML for UML, BPMN, ArchiMate and network diagrams, Vega, Vega-Lite and ECharts for charts, HTML/CSS for cards and layouts. Once installed it answers to the name documd-visuals.
What is needed for the figures to render?
A docu.md-compatible Markdown viewer, because a plain renderer displays the source as a code block. The optional html, epub, docx or pdf export also needs Node 18+ and Chrome/Chromium, with the npx @markdown-viewer/documd CLI installed on its own.
Source text
SKILL.md · MIT
--- name: markdown-viewer-skills description: | One routing skill for text-driven visuals in Markdown, from the docu.md pipeline: charts and analytics (Vega-Lite, ECharts), UML, BPMN, ArchiMate and cloud, network and security architecture (PlantUML), infographics, system-architecture layouts and cards in HTML/CSS. Figures are text in the Markdown, rendered live by the viewer. triggers: - "markdown viewer" - "vega chart" - "PlantUML" - "infographic" - "UML diagram" - "cloud architecture diagram" - "network diagram" das: category: data-visualization upstream: "https://github.com/markdown-viewer/skills" upstream_path: "documd-visuals/SKILL.md" version: latest install: false --- # markdown-viewer-skills > Catalogue stub — full package: [markdown-viewer/skills](https://github.com/markdown-viewer/skills) <!-- das:install:start — generated from stub.yaml by `npm run generate`; edit stub.yaml, not this block --> ## Decision tree 1. **Is the full skill already installed?** Upstream names this skill `documd-visuals`, so it installs next to this pointer (the pointer is not replaced). - Global: `f=~/.agents/skills/documd-visuals/SKILL.md; [ -f "$f" ] && ! grep -q "^das:" "$f" && echo "installed" || echo "pointer"` - Project: `f=.agents/skills/documd-visuals/SKILL.md; [ -f "$f" ] && ! grep -q "^das:" "$f" && echo "installed" || echo "pointer"` - `installed` → the skill is installed as `documd-visuals`, invoke it and proceed - `pointer` → go to step 2 2. **Detect scope, then install:** ```bash [ -e ~/.agents/skills/markdown-viewer-skills ] && echo "global" || echo "project" ``` **Global:** ```bash npx skills add markdown-viewer/skills --skill documd-visuals -g -y ``` **Project:** ```bash npx skills add markdown-viewer/skills --skill documd-visuals -y ``` > **Claude Code:** send a command as a chat message starting with `!` to run it without leaving the conversation. **Heads-up** - **Requires:** A docu.md-compatible Markdown viewer renders the figures live; plain Markdown renderers show them as code blocks; Optional export to html, epub, docx or pdf needs Node 18+ and Chrome/Chromium (npx @markdown-viewer/documd, installed separately) - **License:** CC-BY-4.0 <!-- das:install:end --> ## Invoke after install - Skill name: `documd-visuals` - Trigger phrases: "vega chart", "PlantUML", "infographic", "UML diagram", "cloud architecture diagram", "network diagram" ## What it does Upstream used to ship many separate engine- and domain-named skills (vega, uml, cloud, network, security, archimate, bpmn, mindmap, infographic, architecture, infocard and others). They are now one package, `documd-visuals`, organised by what the reader wants: a router matches the request to a goal domain (service reliability, product metrics, software design, cloud architecture, planning and roadmaps, internal documents and more), goal guides list the scenarios, and engine references cover PlantUML (UML, ArchiMate, BPMN, cloud/network/security architecture with a large icon stencil set), Vega and Vega-Lite, ECharts, Infographic templates, and bare HTML/CSS for architecture diagrams and cards. It ships a library of verified examples to start from. Figures are written as text in the Markdown and drawn live by a docu.md-compatible viewer, so nothing is pre-rendered. Exporting a document to html, epub, docx or pdf is optional and goes through the separate `npx @markdown-viewer/documd` CLI, which this install does not fetch. ## When NOT to use - Slide decks or math notation: the package excludes them - Mermaid, JSON Canvas, draw.io or Graphviz output: the package marks these as not recommended; use `excalidraw-agents365` or `hand-drawn-diagrams` for sketch-style diagrams - Charts rendered as images through a hosted API or as G2/G6 code → use `antvis-chart-skills`
© podo · MIT · podo/design-agent-skills