convez.ai
SkillProdutividadeIntermediário
Hasasasa

HTML-to-PPTX

Hasasasa/claude-skill-html-to-pptx

Converte HTML para PPTX mantendo texto editável e vetorial

Ver no GitHub
BomPluginScore · de 0 a 100

Skill Claude que converte decks HTML em apresentações PowerPoint editáveis, preservando texto como caixas de texto nativas em vez de imagens achatadas. Suporta múltiplos formatos de deck HTML (Reveal.js, beautiful-html-templates) com detecção automática de slides e fallback inteligente para efeitos complexos (gradientes, sombras, filtros).

Instalação

$npx convez-ai Hasasasa/claude-skill-html-to-pptx

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
101
Forks
10
Acompanhando
1
Saúde
42%
Último commit
há 4 meses
Versão
—
Issues abertas
0
Contribuidores
1

Casos de uso

  • Converter apresentações web em PowerPoint sem perder editabilidade
  • Transformar templates HTML em decks profissionais reutilizáveis
  • Preservar hierarquia de conteúdo e semântica textual em PPTX

Veredito

Ferramenta sólida que resolve um problema real: mantém texto editável ao converter HTML para PPTX, diferente de converters que rasterizam tudo. A documentação é clara e o suporte a múltiplos formatos de deck HTML adiciona flexibilidade prática.

README

A tradução para o português deste README ainda está na fila.

Convert HTML slide decks into editable .pptx: text stays as native PowerPoint textboxes, fonts are subset on demand, complex CSS decorations fall back to local snapshots, and HTML/PPT side-by-side audit material is generated for review.

Why this exists

HTML is a great medium for slide decks — layout, typography, animation, and complex visuals are all more flexible than in PowerPoint. But delivery often requires .pptx, and most "HTML → PPT" tools flatten each page into a single image, leaving text uneditable, unsearchable, and pixelated when zoomed.

html-to-pptx avoids the "whole-page screenshot" route. It decomposes HTML into objects PowerPoint understands: text becomes editable textboxes; simple geometry becomes native shapes / lines; effects PPT can't express natively (gradients, shadows, filters, blend modes) are snapshotted locally and placed as a decoration layer behind the vector text.

The result: the output stays close to the HTML source while remaining searchable, editable, and rescalable in PowerPoint / WPS — not a stack of immutable images.

Gets better with use

Each install keeps a private references/lessons-learned.md — seeded once from the committed template, then gitignored. When the audit loop turns up a generalizable HTML anti-pattern or OOXML boundary, the fix recipe gets appended there. Every subsequent convert re-reads the file, so decks sharing traits with ones you've already audited tend to land cleaner on the first pass.

The local copy is never overwritten by upstream — git pull updates the template, not your accumulated notes.

What it handles

Supports single-file HTML decks from beautiful-html-templates / guizang-ppt-skill / Reveal.js / hand-written decks / browser fullscreen pages, with automatic slide detection.

CSS → PPT routing:

HTML element / CSSWhat it becomes in PPT
Text, rich text, color, size, tracking, line-height, alignmentEditable textboxes — searchable and scalable
Background color, border, border-radius, linesNative OOXML geometry
gradient, box-shadow, filter, backdrop-filter, mix-blend-mode, complex transformsLocal snapshot underlay; text still drawn on top as vector
SVG, images, canvasEmbedded directly
Google FontsSubset by characters actually used; CJK content auto-seeds Noto Sans SC / Noto Serif SC

For the full CSS coverage matrix, see references/supported-css.md.

Quick start (Claude Code)

Install

In a Claude Code session, just hand Claude the repo URL:

Install this skill for me: https://github.com/Hasasasa/claude-skill-html-to-pptx

Claude will git clone it into the right skills directory for your platform (macOS / Linux: ~/.claude/skills/; Windows: %USERPROFILE%\.claude\skills\) and walk you through installing Python 3.10+, Playwright, and dependencies. System-level steps will still ask for your confirmation.

Use

After install, ask in plain language:

Convert D:\path\to\deck.html to pptx

Full skill invocation rules, audit mode, and fix discipline are in SKILL.md.

Manual usage (without Claude Code)

If you're not using Claude Code, clone the repo and run the CLI directly.

Install

Requires Python 3.10+ and pip.

git clone https://github.com/Hasasasa/claude-skill-html-to-pptx.git
cd claude-skill-html-to-pptx
pip install -r requirements.txt
python -m playwright install chromium

The visual audit step renders .pptx to PNG. Either of the following works:

  • Windows + Office
  • LibreOffice

requirements.txt already includes pywin32 (Windows-only) and pdf2image. For the LibreOffice route, you also need LibreOffice installed on the system; on Windows, pdf2image typically needs Poppler too. The .pptx itself still generates without a renderer — only the HTML/PPT side-by-side audit images are skipped.

Run

python convert.py path/to/deck.html

Output goes to path/to/deck.pptx next to the input. On first run, a working copy at path/to/deck.audited.html is created automatically; all subsequent audit fixes modify this copy, leaving the original HTML untouched.

Common flags:

FlagWhat it does
--out <path>Custom output .pptx path
--keep-screenshotsKeep per-slide HTML reference screenshots and measurement files
--install-user-fontsInstall non-CJK fonts into the user font directory so WPS / PowerPoint COM can render them
--no-embed-fontsSkip font embedding — smaller file but may fall back to system fonts on other machines
--no-preflightSkip the Stage 1 risk pre-scan
--no-verifySkip the Stage 5a structural self-check
--no-visual-auditSkip the Stage 5b visual audit material generation
--only-slides 2,7,12Re-run only the listed slides (incremental mode for audit iteration)
--cleanupClean up audit / measurement / preflight intermediate artifacts

More demos

Each pair: HTML rendering on the left, html-to-pptx output (rendered in PowerPoint) on the right.

Signal

Market Outlook

OpenClaw — Human-AI Collaboration Framework

8-Bit Orbit

Apex Group (Bold Poster)

Broadside

Pipeline

[1 preflight] -> [2 measure] -> [3 assemble] -> [4 embed fonts] -> [5a self check] -> [5b visual audit]
 preflight.py     measure.py      assemble.py     embed_fonts.py      self_check.py      visual_audit.py
  • preflight: scan for high-risk HTML/CSS patterns.
  • measure: probe the DOM with Playwright, extracting text, shape, media, and decoration records.
  • assemble: write OOXML — textbox / shape / media / snapshot.
  • embed_fonts: subset and embed fonts by characters actually used.
  • self_check: scan PPTX for structural risks and call the renderer to export per-slide PNGs.
  • visual_audit: generate HTML/PPT side-by-side images, contact sheet, audit index, and the review prompt.

Design principles

PrincipleWhat it means
Vector-firstText and simple geometry stay as native PPT objects — editable, searchable, scalable
Local fallbackOnly complex decorations go through the snapshot channel — no whole-page rasterization
Portable fontsGoogle Fonts auto-resolved and subset-embedded to reduce cross-machine drift
AuditablePer-slide HTML/PPT comparison material lets visual issues be located and iterated
Source-preservingA .audited.html working copy is created automatically; the original HTML is never modified

Further reading

Ver o original no GitHub

Bom para

Relacionados