Skip to content
Get packs

Image Generation (Qwen-Image)

Lets the agent generate images via the SiliconFlow Qwen-Image API and save them into your project.

Source Download skill
Repository
xerrors/Yuxi
Stars
7,576
Licence
MIT
Files
1 file

What it does

When you ask for a picture, poster or illustration, the agent writes a short Python script that sends your description to an external image-generation service (SiliconFlow's Qwen-Image model by default), downloads the resulting image inside its sandbox, and saves it to your project folder. It then displays the image in the chat and, if you want a downloadable file, registers it as an artifact.

Great for

  • Non-designers who need a quick illustration or poster draft inside an agent workflow
  • Teams already using SiliconFlow who want Qwen-Image output saved directly into a project
  • Developers who want a template for calling an image-generation API from an agent sandbox

Skip it if

  • Sensitive or confidential prompts, because the description is sent to a third-party API (api.siliconflow.cn) and the file's content leaves your environment
  • Work that needs guaranteed, fully controlled output such as final brand assets without human review, since generated images can be unpredictable
  • Setups without a SILICONFLOW_API_KEY configured in the agent sandbox, where the skill cannot produce images

Try asking

  • “Generate a minimalist poster for a summer music festival and save it to the marketing folder”
  • “Create a watercolor-style illustration of a cat reading a book for my blog post”
  • “Make three variations of a product hero image and show them as a gallery”
Step-by-step guide: how to use Image Generation (Qwen-Image)Install, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 1 original files, the MIT licence and an attribution note.

    image-gen.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip image-gen.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip image-gen.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Describe the subject, style, composition and colours clearly; the agent will use sensible defaults if you give little detail
  • Mention the target folder or file name if you care where the image is saved
  • Check that SILICONFLOW_API_KEY is configured in the sandbox before running, and review the script if you are concerned about the external API call

What's inside

  • SKILL.md

Redistributed unmodified from xerrors/Yuxi under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.