Getting setup

[rigid tundra]

I need the ability to generate worksheets with images how can I setup the image generation easily

[chrome spire]

To help others find answers, you can mark your question as solved via Right click solution message -> Apps -> ✅ Mark Solution

[hushed onyx]

Easiest way in OpenClaw is to enable a bundled image-gen skill and drop in an API key. Two common routes:

Option A (usually the smoothest): nano-banana-pro (Gemini image API)

1. Install the one required binary: uv (Python runner).
2. Add your Gemini key in ~/.openclaw/openclaw.json:

{
  skills: {
    entries: {
      "nano-banana-pro": {
        enabled: true,
        apiKey: "YOUR_GEMINI_API_KEY"
      }
    }
  }

}

3. Start a new chat/session (or restart the gateway), then ask something like:

“Generate a simple black-and-white worksheet illustration of a frog life cycle. No text. Clean line art. White background.”

That skill’s script prints a MEDIA: line, so OpenClaw will typically auto-attach the generated image back into chat.

Docs: https://docs.openclaw.ai/tools/skills
Config: https://docs.openclaw.ai/tools/skills-config

Option B: openai-image-gen (OpenAI Images API)

• Requires python3 + OPENAI_API_KEY set (same idea: put it under skills.entries in openclaw.json).
• It’s more “batch/gallery”-oriented, but works fine if you already have an OpenAI key.

---