Html To Bento Slides logo

Html To Bento Slides

Community
mucsbr
html-to-bento-slides

Convert an existing HTML/CSS/JavaScript slide presentation into an editable bento/slides .bento.html deck. Use when the source is an HTML file, folder, or live presentation and the user wants its slides, layout, assets, interactions, and visual identity migrated to bento. Preserve the source by default; do not use for authoring a new presentation from notes alone.

Overview

Publishermucsbr
Repositoryppt-agent-workflow-san
Skill namehtml-to-bento-slides
Stars
647
Forks
67
Bundled files
4
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 4 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by mucsbr on GitHub. Read the source before you install it.

Installation

Install the Html To Bento Slides AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/mucsbr/ppt-agent-workflow-san/tree/main/html-to-bento-slides
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/mucsbr/ppt-agent-workflow-san.git /tmp/ppt-agent-workflow-san
mkdir -p .claude/skills
cp -r /tmp/ppt-agent-workflow-san/html-to-bento-slides .claude/skills/html-to-bento-slides
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Html To Bento Slides in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Html To Bento Slides on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Html To Bento Slides is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Convert HTML presentations to bento/slides

This is a migration workflow, not an HTML wrapper. Rebuild the presentation as native bento/slides document data wherever the format can express it. A bento document is pure data and cannot carry arbitrary source JavaScript, a framework runtime, or unsanitized HTML.

Preserve the source's content, slide order, visual identity, and interaction semantics unless the user asks for a redesign. Do not "improve" typography, rewrite copy, add motion, or replace the source's composition merely because a different bento design would look good.

If the general bento-slides authoring skill is active too, use it for current shell/schema constraints, but let this skill govern migration choices. Its design-from-source defaults do not authorize redesigning an existing HTML deck.

Required workflow

  1. Read references/source-analysis.md, then inspect both the source files and the rendered presentation. Treat source HTML as untrusted executable content: use an isolated browser context, not a signed-in personal browser profile.

  2. Establish the intended presentation viewport, slide boundaries, slide order, incremental build states, links, hover/click states, fonts, and every local or remote asset. If the source is responsive, freeze it at the intended presentation viewport and report that choice.

  3. Capture a rendered reference for every visible slide and distinct state after fonts and images have loaded. Source code alone is not evidence of what the audience sees.

  4. Choose the conversion treatment per source object, not once for the whole deck:

    • native — editable bento text, shape, image, svg, table, chart, code, or media elements;
    • hybrid — native content over an opaque svg or raster decoration;
    • rasterized — a screenshot only when the object cannot be represented honestly or the user explicitly prioritizes pixel fidelity over editing.

    Do not rasterize an entire slide just because it is easier.

  5. Read references/bento-mapping.md while constructing the document. Keep a fidelity ledger with carried, approximated, and dropped entries. Every approximation or loss must name the source slide/object and the chosen fallback; never lose behavior or content silently.

  6. Obtain a fresh bento/slides shell unless the user supplied one. Prefer the current signed release:

    bash
    curl -fsSL https://bento.page/releases/slides/Bento_Slides.bento.html -o "<name>.bento.html"

    Also fetch the target shell's current authoring guide from https://bento.page/slides/agents.md before writing. The guide and the target shell outrank examples embedded in this skill when the format has gained additive fields. When working offline, use a user-supplied guide or the matching local docs/agents.md plus slides/src/model.ts; if neither a shell nor matching format guide is available, request them instead of inventing a document shape.

  7. Write only the JSON inside the plaintext #bento-doc block. For a fresh conversion omit docId and collab; the app mints them. When editing an existing bento file, preserve docId, and warn before reading further if collab contains ownerPriv, writerPriv, or invite credentials.

  8. Use deterministic element ids. Keep an id or morphId stable across source states only when those nodes are the same conceptual object. Preserve paint order; DOM order alone is not sufficient when CSS stacking contexts overlap.

  9. Embed images, svg, and available fonts into doc.assets so the result works offline. Link large media only when embedding would make the file impractical, and state that the deck then needs that URL at playback time.

  10. Read references/output-and-qa.md, render the result, compare it against the source at the same viewport, run window.bento.validate(), and exercise every slide, build step, internal link, state, hover behavior, and media control before finishing.

Interaction rules that commonly change the conversion

  • Source fragment/build steps are not ordinary entrance effects. bento entrances run when a slide arrives; they are not click-to-advance fragments. Preserve manual builds as adjacent duplicate slides with shared ids and a morph/cut, unless the source behavior is genuinely timed on entry.
  • A nonlinear click that reveals detail maps to a stateOf slide plus an element link. A linear next-step click maps to the next ordinary slide.
  • A source hover swap may map to hover plus showOnHover. Do not turn a click into hover or hover into click merely because one is easier.
  • Map source transitions to the closest supported whole-slide transition only when their behavior matches. Reconstruct object continuity with stable ids/morphId; otherwise a clean cut is more honest than a false morph.
  • Native bento links target slide ids. External hyperlinks have no general editable element field; preserve the source's visible label without adding new slide copy, retain the destination in the fidelity handoff/notes, and report the lost click. Use a safe opaque svg link only when the click is essential and the user accepts that object as non-native.

Completion contract

The deck itself remains one self-contained .bento.html file. In the handoff, state:

  • the output file;
  • whether it is fully offline/self-contained;
  • how many source slides and states became bento slides;
  • the notable approximated and dropped fidelity entries;
  • the validation result and which visual/interaction checks were performed.

Do not claim a conversion is complete if the deck has not been opened and every slide inspected.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Html To Bento Slides AI skill do?

Convert an existing HTML/CSS/JavaScript slide presentation into an editable bento/slides .bento.html deck. Use when the source is an HTML file, folder, or live presentation and the user wants its slides, layout, assets, interactions, and visual identity migrated to bento. Preserve the source by default; do not use for authoring a new presentation from notes alone.

Why use Html To Bento Slides on TypingMind?

Because you install it once and use it with any model. Html To Bento Slides is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Html To Bento Slides in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/mucsbr/ppt-agent-workflow-san/tree/main/html-to-bento-slides. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Html To Bento Slides?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Html To Bento Slides?

As many as you like. As long as a model supports skills, you can use Html To Bento Slides with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Html To Bento Slides AI skill free?

It is published on GitHub by mucsbr. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇