sense-infographic
Skill0 starsChanged 7 months agoMITView on GitHub
Generates infographic images using the SenseNova U1 Fast model, specialized for information-rich visual content. Use this skill when asked to create infographics, information diagrams, data visualizations, flowchart images, process diagrams, or any structured visual representation of information, such as "生成一张信息图" or "制作信息图" or "create an infographic about X" or "generate a data visualization" or "画一个流程图" or "制作知识图谱" or "生成海报" or "信息可视化". Excellent at rendering text-heavy visuals — posters, educational graphics, comparison charts, and step-by-step guides. Also use when the user wants to visualize complex information with Chinese text in the image.
- Reads credentials
- Sends data out
In this repository2 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- text-to-imageSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 2 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 23
Step-0: Check that the environment variable $SENSE_API_KEY exists.
- Very specific: line 44
If error is not null, stop and tell the user the error message.
- Very specific: line 45
If images has entries, download them one by one to the current directory.
- Distinctive: line 46
If the user has explicitly specified a save path, save there instead.
- Very specific: line 83
When the user provides an aspect ratio, convert it to the corresponding size:
- Distinctive: line 111
Include all text verbatim: every label, heading, and caption you want rendered should appear in the prompt exactly as you want it.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Generate infographic images by calling the SenseNova U1 Fast API. This model is purpose-built for information graphics — it handles long structured prompts and renders Chinese text within images exceptionally well.
Execution Steps
Step-0: Check that the environment variable $SENSE_API_KEY exists. Keep it secret. If not set, stop and prompt the user to configure it.
Step-1: Extract parameters from user input:
prompt(required): The infographic description. U1 Fast supports prompts up to 4096 tokens, so be generous with detail — describe the layout, sections, text content, colors, icons, and visual style.size(optional, default:2752x1536): Image dimensions inWIDTHxHEIGHTformat. See Available Sizes.aspect_ratio(optional): Aspect ratio like16:9,9:16,1:1. If provided, convert tosizeusing the Aspect Ratio table. Takes precedence oversize.n(optional, default:1): Number of images to generate.
Step-2: Execute the curl command to generate the image:
RESPONSE=$(curl -s -X POST "https://token.sensenova.cn/v1/images/generations" \
-H "Authorization: Bearer $SENSE_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"sensenova-u1-fast","prompt":"<prompt>","size":"<size>","n":<n>}')
echo "$RESPONSE" | jq -r '{error: .error.message, images: [.data[].url]}'Step-3: Extract the jq result — error is the error message text, images is a list of image URLs from data[].url.
- If
erroris not null, stop and tell the user the error message. - If
imageshas entries, download them one by one to the current directory. - If the user has explicitly specified a save path, save there instead.
Result Example
Successful response
{
"error": null,
"images": [
"https://cdn.sensenova.dev/gen/..."
]
}Failed response — invalid parameter
{
"error": "field Size invalid, should be one of: 1664x2496, 2496x1664, ...",
"images": []
}Failed response — auth failure
{
"error": "Forbidden",
"images": []
}Step-4: Return the file path(s) of the downloaded image(s).
Aspect Ratio
When the user provides an aspect ratio, convert it to the corresponding size:
| Aspect Ratio | Size | Best For |
16:9 | 2752x1536 | Landscape infographic (default) |
9:16 | 1536x2752 | Mobile / vertical infographic |
1:1 | 2048x2048 | Square format |
3:2 | 2496x1664 | Standard landscape |
2:3 | 1664x2496 | Standard portrait |
4:3 | 2368x1760 | Wide landscape |
3:4 | 1760x2368 | Tall portrait |
5:4 | 2272x1824 | Near-square landscape |
4:5 | 1824x2272 | Near-square portrait |
21:9 | 3072x1376 | Ultrawide banner |
9:21 | 1344x3136 | Ultra-tall strip |
Available Sizes
Can also be specified directly in WIDTHxHEIGHT format (lowercase x):
2752x1536, 1536x2752, 2048x2048, 2496x1664, 1664x2496, 2368x1760, 1760x2368, 2272x1824, 1824x2272, 3072x1376, 1344x3136
Prompt Writing Tips
U1 Fast is designed for information graphics and renders text (including Chinese) inside images with high fidelity. To get the best results:
- Describe the layout: specify whether sections flow left-to-right, top-to-bottom, in a grid, or along a timeline.
- Specify color themes: e.g., "以柔和的粉色、淡黄色和浅蓝色为主色调" (soft pink, light yellow, and light blue palette).
- Include all text verbatim: every label, heading, and caption you want rendered should appear in the prompt exactly as you want it.
- Mention visual elements: icons, illustrations, decorative elements (e.g., "配有可爱的拟人化图标").
- State the visual style: flat design, hand-drawn, minimalist, corporate, etc.
The richer and more structured the prompt, the better the output. A good infographic prompt is often 200-800 words long.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

