ClawKit Logo
ClawKitReliability Toolkit
Back to Registry
Official Verified

plume-image

Plume AI image generation and editing service. Auto-triggers when users send images or describe image needs. Supports: text-to-image, image-to-image, background removal, watermark removal, style transfer, text-to-video, image-to-video. Activate when user mentions: generate image, edit image, remove background, change background, remove watermark, text-to-image, image-to-image, AI art, style transfer, generate poster, photo editing, generate video, AI video, animate this image, make it move, image-to-video, make image into video, seedance2, seedance.

skill-install β€” Terminal

Install via CLI (Recommended)

clawhub install openclaw/skills/skills/dake6767/plume-image
Or

Plume AI Image Service

Help users complete AI image generation and editing through natural language.

🚨 Critical Workflow Rules (Must Follow)

All tasks must use: create β†’ poll_cron.py register β†’ reply to user immediately β†’ end

Strictly prohibited:

  • ❌ Do NOT use process_image.py poll command
  • ❌ Do NOT run check after register
  • ❌ Do NOT use sleep to wait before polling
  • ❌ Do NOT ask users to paste API Key in chat
  • ❌ Do NOT auto-create tasks when user sends only an image without text instructions β€” the user may just be preparing a reference image; wait for explicit text instructions before acting

Correct approach:

  • βœ… Call poll_cron.py register immediately after creating a task
  • βœ… Reply to user right after register succeeds, informing them that processing has started
  • βœ… End the current session; background process polls automatically and delivers result
  • βœ… When receiving a pure image (no text), just reply "Got your image. What would you like to do with it?" and wait for the user's next message

⚠️ Mandatory Pre-check (Run Before Every Use)

python3 ${CLAUDE_SKILL_DIR}/scripts/check_config.py
  • Output CONFIGURED: proceed with subsequent operations
  • Output NOT_CONFIGURED: stop immediately, prompt user to configure as follows (do NOT ask for the key in chat):

To use the Plume AI image service, you need to configure your API Key first:

  1. Visit Plume Platform to register and obtain an API Key
  2. Edit ~/.openclaw/openclaw.json, add to skills.entries:
    "plume-image": { "env": { "PLUME_API_KEY": "your-key" } }
    
    Or add PLUME_API_KEY=your-key to the ~/.openclaw/.env file
  3. Enter /restart to apply changes

Background Polling Behavior

This skill starts a background Python process via poll_cron.py register to poll task status. Results are delivered via openclaw message send when the task completes.

  • Poll interval: images 3–5s, videos 10–30s
  • Default timeout: 30 minutes (adjustable by category)
  • Max concurrency: 5 active polling tasks
  • Process exits automatically and cleans up metadata after task completes, fails, or times out
  • Results delivered via openclaw message send, no direct channel API calls

Supported categories (fixed enum, do not invent)

categoryAliasUse caseDefault
Banana2香蕉text-to-image / image-to-image / style transferβœ… default for images
BananaPro香蕉Protext-to-image / image-to-image (user explicitly requested)
remove-bgbackground removal
remove-watermarkwatermark removal
seedream即撦/θ±†εŒ…ε³ζ’¦Doubao Seedream image generation
veotext-to-video / image-to-videoβœ… default for videos
seedance2Seedance2 video (user explicitly requested)

Metadata

Author@dake6767
Stars3376
Views1
Updated2026-03-24
View Author Profile
AI Skill Finder

Not sure this is the right skill?

Describe what you want to build β€” we'll match you to the best skill from 16,000+ options.

Find the right skill
Add to Configuration

Paste this into your clawhub.json to enable this plugin.

{
  "plugins": {
    "official-dake6767-plume-image": {
      "enabled": true,
      "auto_update": true
    }
  }
}
Safety NoteClawKit audits metadata but not runtime behavior. Use with caution.