marp-slide-quality — Claude Code skill: what it does, how to install
marp-slide-quality evaluates Marp Markdown presentations via SlideGauge scoring, detecting layout, accessibility, and density issues directly in source files.
MIT 202 installs ★ 35 updated 2026-10-21
npx skills add podo/design-agent-skills --skill marp-slide-quality -a claude-codeWhat marp-slide-quality does
The skill audits Markdown presentations that include marp: true in their frontmatter. It evaluates slide content across accessibility parameters like image alt attributes and contrast, alongside checks for code snippets, color palettes, content density, and structural layout. After computing an overall score and pass rate, it flags concrete problems like bloated bullet lists or dense paragraphs and offers remediation guidance.
Without uv installed on the host system, the skill cannot run: it fetches and executes SlideGauge from GitHub at runtime via uvx, rather than relying on a local PyPI package. It also relies on jq for JSON parsing tasks. It processes source Markdown rather than exported HTML or PDF slides, and it does not handle Reveal.js or Slidev decks.
What it helps with
- audit Marp slide source files for contrast and alt-text gaps
- spot overly dense text blocks and excessive bullet lists
- score presentation layout and color balance before exporting
- verify slide deck fixes against SlideGauge quality metrics
Example prompts for the agent
- Run marp slide quality on
presentation.mdto see my score - Check this deck with SlideGauge and suggest fixes for layout issues
- Use marp-slide-quality to review contrast and content density
How to install marp-slide-quality
1Skills CLI (Vercel)
npx skills add podo/design-agent-skills --skill marp-slide-quality -a claude-code2Manually: put the files in place
~/.claude/skills/marp-slide-quality/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/marp-slide-quality/
1Skills CLI (Vercel)
npx skills add podo/design-agent-skills --skill marp-slide-quality -a codex2Manually: put the files in place
.agents/skills/marp-slide-quality/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add podo/design-agent-skills --skill marp-slide-quality -a cursor2Manually: put the files in place
.cursor/skills/marp-slide-quality/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install podo/design-agent-skills skills/marp-slide-quality2Manually: put the files in place
.github/skills/marp-slide-quality/SKILL.md1Skills CLI (Vercel)
npx skills add podo/design-agent-skills --skill marp-slide-quality -a gemini-cli2Platform CLI
gemini skills install https://github.com/podo/design-agent-skills --path skills/marp-slide-qualityManually: put the files in place
.devin/skills/marp-slide-quality/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add podo/design-agent-skills --skill marp-slide-quality -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 marp-slide-quality
How to install marp-slide-quality in Claude Code?
Run npx skills add podo/design-agent-skills --skill marp-slide-quality -a claude-code in your terminal.
What does marp-slide-quality do?
It scores Marp presentation Markdown files using SlideGauge, pinpointing problems across accessibility, code, palette, layout, and text density.
What dependencies are required to run this skill?
You need uv to provide the uvx command for fetching SlideGauge on demand, an active internet connection, and jq for handling JSON filters.
Source text
SKILL.md · MIT
--- name: marp-slide-quality description: | Analyzes and improves Marp Markdown presentations using SlideGauge quality scoring — checks accessibility (alt text, contrast), code, color, content density and layout on the Markdown source. Use as a companion to marp-slides, run after slide creation. triggers: - "marp slide quality" - "SlideGauge" - "marp-slide-quality" - "nibzard marp" das: type: skill category: presentations upstream: "https://github.com/nibzard/marp-slide-quality" upstream_path: "SKILL.md" version: latest install: true --- # marp-slide-quality > Catalogue stub — full skill: [nibzard/marp-slide-quality](https://github.com/nibzard/marp-slide-quality) <!-- 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?** - Global: `f=~/.agents/skills/marp-slide-quality/SKILL.md; [ -f "$f" ] && ! grep -q "^das:" "$f" && echo "installed" || echo "pointer"` - Project: `f=.agents/skills/marp-slide-quality/SKILL.md; [ -f "$f" ] && ! grep -q "^das:" "$f" && echo "installed" || echo "pointer"` - `installed` → the skill is installed as `marp-slide-quality`, invoke it and proceed - `pointer` → go to step 2 2. **Detect scope, then install:** ```bash [ -e ~/.agents/skills/marp-slide-quality ] && echo "global" || echo "project" ``` **Global:** ```bash npx skills add nibzard/marp-slide-quality --skill marp-slide-quality -g -y ``` **Project:** ```bash npx skills add nibzard/marp-slide-quality --skill marp-slide-quality -y ``` > **Claude Code:** send a command as a chat message starting with `!` to run it without leaving the conversation. **Heads-up** - **Requires:** uv (provides uvx): the skill runs `uvx --from git+https://github.com/nibzard/slidegauge slidegauge ...`, which fetches and executes SlideGauge from that GitHub repo at use time (not a PyPI package); jq for the JSON-filter examples in the skill - **License:** Apache-2.0 <!-- das:install:end --> ## Invoke after install - Skill name: `marp-slide-quality` - Trigger phrases: "marp slide quality", "SlideGauge", "marp-slide-quality", "nibzard marp" ## What it does Runs SlideGauge quality analysis on Marp Markdown presentations (files with `marp: true` frontmatter) and helps fix what it finds: overall score and pass rate, plus per-category issues for accessibility, code, color, content and layout. Produces a scored report with specific fix recommendations (too many bullets, overlong content and similar) and re-runs the analysis after changes. Best used as a post-creation review pass after `marp-slides` generates the initial deck. SlideGauge is not bundled: the skill runs it through `uvx` from its own GitHub repository each time, so it needs `uv` and network access, and it executes code fetched from that repository. ## When NOT to use - Reveal.js or Slidev presentations — SlideGauge is Marp-specific - Creating slides from scratch — use `marp-slides` first, then run this for quality review - When you need Marp rendered output — this analyzes the Markdown source, not the render
© podo · MIT · podo/design-agent-skills