Guide · Agent skill
How to use the Infographic Generator agent skill (baoyu-infographic)
Updated · written with AI from the project's own documentation and checked against our catalogue data
The baoyu-infographic skill teaches an AI agent to turn a piece of text, such as a report, article or dataset, into a finished infographic image. It picks one of 21 information layouts and one of 22 visual styles, then hands a prompt to an image generation tool you already have. This guide covers what the skill does, how to install it, how to ask for an infographic, and what it needs to run.
Infographic GeneratorJimLiu/baoyu-skills · MIT Download the skill Quick start
- 1Download and unzip. Download the skill ZIP from this page and unzip it into ~/.claude/skills/ or a project's .claude/skills/ folder.
- 2Check your image backend. Confirm you have Codex imagegen, Cursor GenerateImage, or an installed image skill such as baoyu-image-gen available.
- 3Make your request. Paste your text or give a file path and ask for an infographic, stating the audience and the main takeaway.
- 4Confirm the recommendation. Review the proposed layout, style, aspect ratio, language and backend, then reply to approve or change them.
- 5Check the prompt file. Open the prompt saved under prompts/ if you want to verify wording before the image is generated.
- 6Review the image. Check every figure against your source. If text is wrong, ask the agent to regenerate from a corrected prompt.
What an agent skill is and what this one does
An agent skill is a folder of instructions that an AI agent loads when your request matches the skill's description. This one is called baoyu-infographic, and it is triggered when you ask for an infographic, a visual summary, a 信息图, a 可视化 or a 高密度信息大图.
Once it is active, the agent analyses your material, recommends a combination of one information layout and one visual style, and produces an infographic image. Layout controls how information is structured; style controls the look. You can combine any layout with any style.
- Analyses your text or material before designing anything
- Recommends a layout × style pair and explains the choice
- Writes the full image prompt to a file under prompts/ before generating
- Asks you to confirm before generation by default
Install the baoyu-infographic skill
Download the ZIP for this skill from this page and unzip it. Put the resulting folder in one of two places: ~/.claude/skills/ for all your projects, or .claude/skills/ inside a specific project folder.
If you use the Claude apps instead, open Settings, go to Capabilities, then Skills, and upload the skill there.
- Personal install: ~/.claude/skills/
- Project install: .claude/skills/ inside your project
- Claude apps: Settings → Capabilities → Skills, then upload
Requirements: an image generation backend
The skill does not draw the image itself with code. It needs an image generation backend in your tool. The source lists three routes: Codex imagegen, Cursor's GenerateImage tool, or a separate image skill such as baoyu-image-gen.
If you run the Codex CLI, the codex-cli route needs the codex command on your PATH with an active login. An optional EXTEND.md preferences file can save defaults such as your preferred image backend. The skill uses the network, but it runs no scripts of its own.
- Codex imagegen, Cursor GenerateImage, or an installed image generation skill such as baoyu-image-gen
- Optional: codex CLI with active login, for the codex-cli route
- Optional: EXTEND.md preferences file for saved defaults
- Runs scripts: no. Uses network: yes.
How to trigger the skill with a natural request
You do not need special syntax. Describe the material and say you want an infographic. Naming the skill directly also works. Give the agent the text or a file path, and say what the viewer should take away.
The agent will propose a layout and style and ask you to confirm the combination, aspect ratio, language and backend before it generates anything. You can reply to change any of these.
- "Make an infographic from this quarterly report: [paste text]"
- "Turn this article into a visual summary in portrait format, in English"
- "Create a 信息图 from this dataset, layout comparison, style flat"
- "Use the baoyu-infographic skill on /path/to/notes.md"
Options you can ask for
The skill accepts a few options. Layout defaults to bento-grid and style defaults to craft-handmade. Aspect can be named (landscape 16:9, portrait 9:16, square 1:1) or a custom ratio such as 3:4 or 2.35:1. Language accepts values such as en, zh or ja.
You can also supply reference images. They can guide style, palette or composition. Use them when you want the result to match an existing brand look.
- --layout: 21 options, default bento-grid
- --style: 22 options, default craft-handmade
- --aspect: landscape, portrait, square or a custom W:H ratio
- --lang: en, zh, ja and others
- --no-confirm: skip confirmation only when you explicitly ask for direct generation
- --ref: reference image files
--layout bento-grid --style craft-handmade --aspect portrait --lang enReview the prompt file and the output
Before generation, the skill writes each image's full prompt to a file under prompts/, named NN-{type}-[slug].md. This file is the record of what was requested, and it lets you switch backends without rewriting the prompt.
The skill also forbids repairing text on the image afterwards. If a label is wrong, the agent should regenerate from a corrected prompt, choose a layout with less on-image text, or ask you which candidate to keep.
- Check the prompt file in prompts/ before generation if you need exact wording
- Expect to regenerate rather than edit text in the image
- Check every number against your source, because AI image text can be wrong
What people use it for
- Turning a long report into a one-page dense visual summary
- Making a shareable infographic from an article for social media or a newsletter
- Explaining a dataset or comparison to students with a comparison layout
- Creating a funnel or dashboard style visual for a marketing presentation
Frequently asked questions
Is the baoyu-infographic skill free?
The skill is published under the MIT licence, as stated on this page. Image generation may depend on the backend you use, such as Codex, Cursor or an image skill, so check the terms of that tool.
What do I need to run baoyu-infographic?
You need an image generation backend: Codex imagegen, Cursor GenerateImage, or an installed image generation skill such as baoyu-image-gen. The codex-cli route also needs the codex CLI with an active login. Network access is used.
Where do I install an agent skill like this?
Unzip the downloaded folder into ~/.claude/skills/ for personal use or into .claude/skills/ inside a project. In the Claude apps, go to Settings, Capabilities, Skills and upload it.
Does the skill generate SVG or editable vector infographics?
No. The skill forbids code-based rendering such as SVG, HTML or canvas as a substitute for a raster image. If no image backend is available, it will ask you how to proceed.
Can I fix a typo in the generated infographic myself?
The skill forbids painting over or repairing text on a generated bitmap with tools such as ImageMagick or Pillow. Regenerate from a corrected prompt instead, or simplify the layout to reduce on-image text.
Does the skill run scripts on my computer?
The source lists runs scripts as no and uses network as yes. It hands the prompt to an image backend rather than running its own scripts.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
Use this skill if you want a dense infographic image from a report, article or dataset and already have Codex, Cursor or an image generation skill set up. Pick something else if you need editable vector or SVG output, or if every figure on the image must be exact, since AI-generated text can be wrong and the skill will not repair it.
Download the skill