Baoyu Imagine

CLI image generation across 10 image APIs with aspect ratios, refs and batch mode

Works with setup

Test report

Verdict
Works with setup
Score
7.2/10
Tested
Jul 21, 2026
Environment
Claude Code 2.x (agent harness)
Upstream re-checked
Aug 10, 2026 · 06828a9

Cloned the repo to a temp dir and installed skills/baoyu-imagine into a throwaway HOME ($(mktemp -d)/.claude/skills) — SKILL.md landed correctly; frontmatter has name+description+version, and raw fetches of scripts/main.ts, references/usage-examples.md and references/config/first-time-setup.md all returned HTTP 200. `bun scripts/main.ts --help` ran and printed an option list that matches the SKILL.md table (same 10 providers, --jobs default 10, quality default 2k, Replicate --n must be 1). I could NOT measure output: with every provider key unset the actual run `bun scripts/main.ts --prompt "A red cube on white background" --image out.png` exited 1 with "No API key found. Set GOOGLE_API_KEY, GEMINI_API_KEY, OPENAI_API_KEY, ..." and wrote no file, so no baseline-vs-skill image comparison exists. `bun test` in the skill dir gave 83 pass / 5 fail, all five failures in build-batch.test.ts from the harness spawning `bun --import tsx` with tsx absent ("Cannot find module './cjs/index.cjs'"); no curl|sh, base64 blobs or secret exfiltration found — the only child_process use is execFileSync("curl", args) in providers/google.ts posting to the Google API with an argv array (no shell).

Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works

  • Installs cleanly 5/5
  • Triggers reliably 5/5
  • Output vs. baseline 4/10
  • Docs & honesty 4/5

What Baoyu Imagine does

Generates images from Claude Code by calling a bundled Bun/TypeScript CLI that wraps ten image APIs: OpenAI GPT Image 2, Azure OpenAI, Google Gemini, OpenRouter, DashScope, Z.AI, MiniMax, Jimeng, Seedream and Replicate. Supports text-to-image, reference-image editing with identity-preservation prompting guidance, aspect ratio and quality presets, and parallel batch generation from saved prompt files. Triggers when the user asks to generate, create or draw an image, or to batch-render a set of prompt files.

How to install Baoyu Imagine

git clone --depth 1 https://github.com/guanyang/open-agent-hub.git /tmp/baoyu-imagine-src
mkdir -p ~/.claude/skills
cp -R /tmp/baoyu-imagine-src/skills/baoyu-imagine ~/.claude/skills/baoyu-imagine
# Requires `bun` on PATH (brew install oven-sh/bun/bun) or it falls back to `npx -y bun`.
# Requires at least ONE provider API key, otherwise the CLI exits 1 with "No API key found".
#   export OPENAI_API_KEY=...   # or GOOGLE_API_KEY / GEMINI_API_KEY / OPENROUTER_API_KEY /
#                               # DASHSCOPE_API_KEY / ZAI_API_KEY / MINIMAX_API_KEY /
#                               # REPLICATE_API_TOKEN / ARK_API_KEY / AZURE_OPENAI_API_KEY+AZURE_OPENAI_BASE_URL
# Keys can instead live in ~/.baoyu-skills/.env or <cwd>/.baoyu-skills/.env
# First run also writes a preferences file (EXTEND.md) via an interactive setup question.
# Manual run: bun ~/.claude/skills/baoyu-imagine/scripts/main.ts --prompt "A cat" --image cat.png --ar 16:9

Skills live in ~/.claude/skills/ (global) or .claude/skills/ (per-project). Restart Claude Code after installing.

Commands — how to trigger Baoyu Imagine

  • /baoyu-imagine CLI image generation across 10 image APIs with aspect ratios, refs and batch mode

It also activates on plain-language prompts like these:

  • Generate an image of a mountain landscape at sunset
  • Create a batch of images from these saved prompts
  • Generate a reference-image edit using GPT Image

Frequently asked questions

Is the Baoyu Imagine skill free?
Yes. The skill itself is free from guanyang/open-agent-hub. SkillProof publishes the install command and an independent test verdict at no cost.
Does Baoyu Imagine work with Claude Code?
We tested it with Claude Code 2.x (agent harness) on Jul 21, 2026. Verdict: Works with setup. Cloned the repo to a temp dir and installed skills/baoyu-imagine into a throwaway HOME ($(mktemp -d)/.claude/skills) — SKILL.md landed correctly; frontmatter has name+description+version, and raw fetches of scripts/main.ts, references/usage-examples.md and references/config/first-time-setup.md all returned HTTP 200. `bun scripts/main.ts --help` ran and printed an option list that matches the SKILL.md table (same 10 providers, --jobs default 10, quality default 2k, Replicate --n must be 1). I could NOT measure output: with every provider key unset the actual run `bun scripts/main.ts --prompt "A red cube on white background" --image out.png` exited 1 with "No API key found. Set GOOGLE_API_KEY, GEMINI_API_KEY, OPENAI_API_KEY, ..." and wrote no file, so no baseline-vs-skill image comparison exists. `bun test` in the skill dir gave 83 pass / 5 fail, all five failures in build-batch.test.ts from the harness spawning `bun --import tsx` with tsx absent ("Cannot find module './cjs/index.cjs'"); no curl|sh, base64 blobs or secret exfiltration found — the only child_process use is execFileSync("curl", args) in providers/google.ts posting to the Google API with an argv array (no shell).
What is the Baoyu Imagine SkillProof Score?
7.2/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 4/10, docs & honesty 4/5.
How do I install Baoyu Imagine?
Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
Can I use Baoyu Imagine with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.