What the tech

Search

Browse topics

/ opens search from anywhere. Esc closes it.

Tools
What the tech
Get the newsletter

baoyu-image-gen: a Claude Code skill by Jim Liu 宝玉

For people who want Claude to create images through their own keys for OpenAI, Google, Replicate and others. Skip it if you have no image API key or Codex login.

High-risk warning At least one scanner rated it high risk, so we keep it out of the top ten. Read its code before you install it.

Copy install command View the source
Score
71of 100
Rank
#165of 239
Installs
34.2Kon skills.sh
Security
Warning 5 scanners
Updated
2026
Context
~117tokens always

Install

npx skills add https://github.com/jimliu/baoyu-skills --skill baoyu-image-gen

Run it in your terminal. View the source

Using the Claude app rather than Claude Code? You can add a skill there too. How to add it in the app

What does baoyu-image-gen do?

baoyu-image-gen is a skill that generates images by running a TypeScript script with Bun, routed to one of a dozen providers such as OpenAI, Google, Azure, OpenRouter, DashScope, MiniMax, Seedream or Replicate. Before the first image it asks for your provider, model, quality and save location and stores them in an EXTEND.md file. Each run takes a prompt or prompt files, an output path, an aspect ratio and optional reference images. It names the provider and model before generating, retries a failed image up to three times, and runs a batch file in parallel when several prompts are saved. A Codex login can stand in for an OpenAI key.

When should you use baoyu-image-gen?

Use it when

  • You need a cover or illustration from a text prompt
  • You have a folder of saved prompts to render at once
  • You want one person kept consistent from reference photos

Skip it when

  • You have no API key for any supported provider
  • You want hand edits inside an image editor

What it needs and what it can touch

You need

  • Bun, or npx to run it
  • An API key for at least one provider (OpenAI, Google, Replicate, DashScope and others), or a logged-in Codex CLI

What it can do

  • Runs shell commands (bun scripts/main.ts)
  • Writes image files and an EXTEND.md preferences file
  • Sends prompts and reference images to the chosen image provider

From reading its instructions, not from running it.

Why it scores 71

Strongest on upkeep (100 of 100), weakest on adoption (51).

  • Adoption35 points 51
  • Upkeep20 points 100
  • Trust20 points 54
  • Quality15 points 100
  • Reader answers10 points Not enough answers yet, so this does not count

How the formula works

Is baoyu-image-gen safe to install?

At least one scanner rated it high risk, so we keep it out of the top ten. Read its code before you install it. Scanners often miss what other scanners catch, and a skill can change after a scan. Read its source before you install it.

ScannerResultDate
Gen Agent Trust HubPassedNov 2, 2026
SocketMedium riskNov 2, 2026
SnykMedium riskNov 2, 2026
RunlayerHigh riskMar 14, 2026
ZeroLeaksLow riskApr 16, 2026

Details

Installs
34,225
Repository stars
27,438
Last repository change
Tokens always loaded
about 117
Tokens when used
about 3,965
License
MIT
Data confidence
High

Alternatives to baoyu-image-gen

All skills for this job

  • mmx-cli

    For developers who want MiniMax models for chat, images, video, speech and transcription from the terminal.

    Score 76 32.5K installs Official

    Get mmx-cli

  • design

    For founders and designers who want logos, brand mockups, banners, icons and social images from one skill.

    Score 74 49.9K installs

    Get design

  • Higgsfield

    For marketers and creators with a Higgsfield account who want images, video, ads or sites made from chat.

    Score 83 24K installs Official

    Get Higgsfield

  • baoyu-infographic

    For writers and marketers who want an article or notes turned into a single infographic image.

    Score 73 33.5K installs

    Get baoyu-infographic