agentleFS
alphaqiualphaqiu/bifang-skills/skills/sense-infographic/SKILL.md

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

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
  1. Very specific:

    Step-0: Check that the environment variable $SENSE_API_KEY exists.

    line 23
  2. Very specific:

    If error is not null, stop and tell the user the error message.

    line 44
  3. Very specific:

    If images has entries, download them one by one to the current directory.

    line 45
  4. Distinctive:

    If the user has explicitly specified a save path, save there instead.

    line 46
  5. Very specific:

    When the user provides an aspect ratio, convert it to the corresponding size:

    line 83
  6. Distinctive:

    Include all text verbatim: every label, heading, and caption you want rendered should appear in the prompt exactly as you want it.

    line 111

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

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 in WIDTHxHEIGHT format. See Available Sizes.
  • aspect_ratio (optional): Aspect ratio like 16:9, 9:16, 1:1. If provided, convert to size using the Aspect Ratio table. Takes precedence over size.
  • 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 error is not null, stop and tell the user the error message.
  • If images has 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 RatioSizeBest For
16:92752x1536Landscape infographic (default)
9:161536x2752Mobile / vertical infographic
1:12048x2048Square format
3:22496x1664Standard landscape
2:31664x2496Standard portrait
4:32368x1760Wide landscape
3:41760x2368Tall portrait
5:42272x1824Near-square landscape
4:51824x2272Near-square portrait
21:93072x1376Ultrawide banner
9:211344x3136Ultra-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.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.