claude-skills/markdown-html/README.md
Claude b99505446a
fix: resolve 11 P0 correctness defects from newgen audit
- fda-consultant-specialist: QMSR transition (eff. 2026-02-02) correctly framed;
  legacy QSR sections mapped to ISO 13485:2016 clauses (17-row mapping)
- risk-management-specialist: ALARP-with-cost-benefit replaced by AFAP per
  EU MDR Annex I GSPR 1-4 + EN ISO 14971:2019/A11
- eu-ai-act-specialist: Art. 5(1)(f) sample corrected (workplace, not retail);
  both dual-published copies kept byte-identical
- financial-analyst: 4 scripts now read bundled sample shape (and flat shape),
  exit non-zero on missing data; verified real outputs (ROE 25%, DCF $13.90/sh)
- deal-desk: fixed-COGS margin math unified across script/reference/SKILL.md;
  margin scoring recalibrated (30% discount on 80% GM now scores 25, not 37)
- senior-qa/backend/frontend + email-template-builder: 8 corrupted code
  literals from past bulk YAML-quoting sweep reconstructed
- google-workspace-cli: install coordinates fixed to verified upstream
  (@googleworkspace/cli); phantom subcommands replaced with verified syntax
- project-management: fabricated Atlassian MCP tool names replaced with the 30
  real Remote MCP tools; new canonical references/atlassian-mcp-tools.md;
  unavailable capabilities routed to UI/REST; orphaned scripts wired
- command-guide: retired (documented a different repo's ecosystem)
- markdown-html: stale v2.10.0 'foundation' text replaced; routing now targets
  the shipped converters; never hand-render
- webinar_funnel_scorer.py: argparse added (--help/--sample/stdin)
- handoff redaction linter: missing 17th pattern (private CIDR) added

https://claude.ai/code/session_019AJddAL1NADWMXsy1qNPQF
2026-06-10 14:32:43 +00:00

113 lines
5.7 KiB
Markdown

# markdown-html — Markdown to interactive HTML converter
> "I generated a 200-line implementation plan in Claude Code last month. Markdown. … I skimmed the first 20 lines, scrolled past the diagram I could not actually parse, and closed the file. I never read the rest."
> — Thariq Shihipar, *Claude Code HTML output* (Medium, 2026)
Convert long markdown files in a Claude project into single-file, lightly-interactive HTML that respects your brand. One-time design-system onboarding captures brand primary + accent + typography + layout style + default save location. Every conversion reads that config and renders consistently.
## Status — v2.10.3 (domain complete)
| Skill | Purpose | Status |
|---|---|---|
| `markdown-html-orchestrator` | Routes long markdown → converter sub-skill (`context: fork`) | ✓ live |
| `design-system` | Onboarding wizard + WCAG-AA-validated brand palette + shared config | ✓ live |
| `md-document` | Long-form: sticky TOC + collapsibles + search + code-copy + scrollspy | ✓ live |
| `md-review` | Code review: 2-col diff + severity-tagged margin annotations + jump-nav | ✓ live |
| `md-slides` | Slide deck: arrow-key nav + presenter mode + print-to-PDF | ✓ live |
## Quick start
```bash
# 1. One-time onboarding (10 questions, ~60 seconds)
python3 markdown-html/skills/design-system/scripts/onboard.py
# 2. (or zero-touch defaults for first-test)
python3 markdown-html/skills/design-system/scripts/onboard.py --defaults
# 3. Inspect the saved config
python3 markdown-html/skills/design-system/scripts/config_loader.py --status
# 4. Classify a markdown file and see the routing decision
python3 markdown-html/skills/markdown-html-orchestrator/scripts/doctype_classifier.py \
--input ./my-report.md --output json \
| python3 markdown-html/skills/markdown-html-orchestrator/scripts/route_explainer.py
# 5. Resolve where it would save
python3 markdown-html/skills/markdown-html-orchestrator/scripts/output_path_resolver.py \
--input ./my-report.md --doctype document
```
## Slash commands
- `/cs:markdown-html <path>.md` — top-level router (classify + route + recommend)
- `/cs:grill-markdown-html <path>.md` — Matt-style 5-question grill before conversion
- `/cs:design-system` — surface the onboarding wizard
- `/cs:md-document <path>.md` — long-form converter
- `/cs:md-review <path>.md` — code-review converter
- `/cs:md-slides <path>.md` — slide-deck converter
## Hard rules
1. **Refuses input < 100 lines** — markdown still wins below the threshold (Shihipar). Keep short docs as markdown.
2. **Refuses without onboarding** — without a captured brand, output renders with placeholder defaults. Run onboarding once.
3. **Refuses unwritable save location** — onboarding asks where to save; the orchestrator honors it; collisions get suffixed (`-2`, `-3`, …).
4. **Single-file HTML only** — all CSS + JS inline. The only external CDN entries are Google Fonts + Prism.js. No build step, no bundler, no JS framework.
5. **Never silently chain** — "convert this AND make slides AND a code review" is three operations, asked explicitly.
6. **WCAG AA enforced** — body-text contrast must reach 4.5:1. Onboarding refuses any combination that fails.
## Distinct from
- **Anthropic Playground plugin** (`/playground`) — builds interactive prompt-tuning controls (sliders, knobs, prompt-copy-back). Different tool entirely.
- **`marketing/landing/`** — generates landing pages from scratch. Doesn't take markdown input.
- **`engineering/handoff/` + `productivity/handoff/`** — session continuity briefs. Different artifact type.
## File layout
```
markdown-html/
├── .claude-plugin/
│ └── plugin.json
├── agents/
│ └── cs-markdown-html-orchestrator.md
├── commands/
│ ├── cs-markdown-html.md # router
│ ├── cs-design-system.md # onboarding surface
│ ├── cs-grill-markdown-html.md # 5-question grill
│ ├── cs-md-document.md # long-form converter
│ ├── cs-md-review.md # code-review converter
│ └── cs-md-slides.md # slide-deck converter
└── skills/
├── markdown-html-orchestrator/ # context: fork
│ ├── SKILL.md
│ ├── scripts/
│ │ ├── doctype_classifier.py
│ │ ├── route_explainer.py
│ │ └── output_path_resolver.py
│ └── references/
│ ├── information_density_canon.md
│ ├── orchestrator_routing_patterns.md
│ └── single_file_html_discipline.md
├── design-system/
│ ├── SKILL.md
│ ├── scripts/
│ │ ├── onboard.py
│ │ ├── config_loader.py
│ │ └── brand_palette_validator.py
│ ├── references/
│ │ ├── design_token_canon.md
│ │ ├── wcag_accessibility.md
│ │ └── typography_pairing.md
│ └── assets/
│ └── design_system_schema.json
├── md-document/ # long-form: markdown_parser → html_renderer → interactivity_injector
├── md-review/ # code review: diff_parser → annotation_extractor → review_html_renderer
└── md-slides/ # slide deck: slide_splitter → presenter_notes_parser → deck_html_renderer
```
## License
MIT. See repo `LICENSE`.
## Spec
Thariq Shihipar — *Claude Code HTML output: Why Markdown Lost and How to Switch* (Medium, 2026). The plugin operationalizes the article's central claim — markdown collapses past ~100 lines for agent-generated artifacts; HTML restores density, clarity, shareability, and lightweight interaction.