Videoagent Image Studio

🎨 VideoAgent Image Studio

Use when: User asks to generate, draw, create, or make any kind of image, photo, illustration, icon, logo, or artwork.

Generate images with 8 state-of-the-art AI models. This skill automatically picks the best model for the job and handles all the complexity — including Midjourney's async polling — so you can focus on the conversation.

---

Quick Reference

---

How to Generate an Image

Step 1 — Enhance the prompt

Before calling the script, expand the user's prompt with style, lighting, and quality descriptors appropriate for the chosen model.

• Midjourney: Add cinematic lighting, ultra detailed, --v 7, --style raw
• Flux: Add masterpiece, highly detailed, sharp focus, professional photography
• Ideogram: Be explicit about text content, font style, and layout
• Recraft: Specify vector illustration, flat design, icon style

Step 2 — Run the script

node {baseDir}/tools/generate.js \
  --model <model_id> \
  --prompt "<enhanced prompt>" \
  --aspect-ratio <ratio>

All parameters:

Step 3 — Return the result

The script always waits and returns the final image URL(s). No polling required.

{
  "success": true,
  "model": "flux-pro",
  "imageUrl": "https://...",
  "images": ["https://..."]
}

Send the imageUrl to the user.

---

Midjourney Actions

After generating a 4-image grid with Midjourney, offer the user these options:

# Upscale image #2 (subtle, preserves details)
node {baseDir}/tools/generate.js \
  --model midjourney \
  --action upscale \
  --index 2 \
  --job-id <job_id>

# Create a strong variation of image #3
node {baseDir}/tools/generate.js \
  --model midjourney \
  --action variation \
  --index 3 \
  --job-id <job_id> \
  --variation-type 1

# Regenerate with same prompt
node {baseDir}/tools/generate.js \
  --model midjourney \
  --action reroll \
  --job-id <job_id>

Upscale types: 0 = Subtle (default, best for photos), 1 = Creative (best for illustrations)

Variation types: 0 = Subtle (default), 1 = Strong (dramatic changes)

---

Example Conversations

User: "Draw a snow leopard on a snowy mountain with cinematic lighting"

# Choose midjourney for artistic quality
node {baseDir}/tools/generate.js \
  --model midjourney \
  --prompt "a majestic snow leopard on a snowy mountain peak, cinematic lighting, dramatic atmosphere, ultra detailed --ar 16:9 --v 7" \
  --aspect-ratio 16:9

🎨 Done! Which one to upscale? (U1-U4) Or create a variant? (V1-V4)

---

User: "Use Flux to generate a perfume product poster, white background"

# Choose flux-pro for photorealistic product shots
node {baseDir}/tools/generate.js \
  --model flux-pro \
  --prompt "a luxury perfume bottle on a clean white background, professional product photography, soft shadows, 8k, highly detailed" \
  --aspect-ratio 3:4

---

User: "Show me a quick draft"

# flux-schnell for instant previews
node {baseDir}/tools/generate.js \
  --model flux-schnell \
  --prompt "..." \
  --aspect-ratio 1:1

---

User: "Make me an App icon, flat style, blue theme"

# recraft for vector/icon style
node {baseDir}/tools/generate.js \
  --model recraft \
  --prompt "a minimal flat design app icon, blue color scheme, simple geometric shapes, vector style, white background"

---

Setup

Zero API keys needed! All requests go through a hosted proxy that handles authentication server-side.

The skill works out of the box — just install and use.

Advanced: Custom proxy or token

If you want to use your own proxy or a persistent token, set these environment variables:

{
  "skills": {
    "entries": {
      "videoagent-image-studio": {
        "enabled": true,
        "env": {
          "IMAGE_STUDIO_PROXY_URL": "https://your-proxy.vercel.app",
          "IMAGE_STUDIO_TOKEN": "your_token_here"
        }
      }
    }
  }
}

To deploy your own proxy, see the videoagent-audio-studio proxy as a reference implementation. You'll need FAL_KEY and LEGNEXT_KEY as Vercel environment variables.

---

Changelog

v2.0.0

• Simplified async: The script now blocks until Midjourney completes. No more --async / --poll flags needed in SKILL.md instructions.
• Unified output format: All models return the same { success, imageUrl, images } shape.
• Reference images for Nano Banana: Pass --reference-images "url1,url2" for character/style consistency across generations.

v1.3.0

• Added non-blocking async mode for Midjourney (--async + --poll).

v1.2.0

• Midjourney turbo mode enabled by default (~10-20s).

v1.1.0

• Switched Midjourney provider from TTAPI to Legnext.ai for better stability.

v1.0.0

• Initial release with Midjourney, Flux, SDXL, Nano Banana, Ideogram, Recraft.

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[2.0.0] - 2026-03-03

Changed

• Simplified SKILL.md: Removed the complex 3-step async/poll workflow from the main instructions. The script already handles all polling internally — the SKILL.md now reflects this with a single, clean "run and get result" pattern.
• Unified output format: All models now return a consistent { success, model, imageUrl, images, jobId } shape, making it easier to handle results uniformly.
• Clearer model selection table: Added "Speed" column so agents can make better trade-off decisions.
• Added "Use when" trigger: SKILL.md now starts with a clear activation condition so the agent knows exactly when to invoke this skill.
• Documented `--reference-images` for Nano Banana: Pass comma-separated URLs for character/style consistency across sequential image generations.

---

[1.3.0] - 2026-02-25

Added

• Non-blocking async mode for Midjourney (--async flag). Submit a job and return immediately with job_id, without waiting for completion. This prevents the bot from being blocked while waiting for image generation.
• Status poll mode (--poll --job-id <id>). Check job status once and return immediately — no waiting. Returns status: "completed", "pending", "processing", or "failed".
• Updated SKILL.md with mandatory async workflow documentation. All Midjourney requests should now use --async + periodic --poll to avoid blocking the bot.

Changed

• --async flag is supported for all Midjourney actions: imagine, upscale, variation, reroll.

---

[1.2.0] - 2026-02-25

Changed

• Midjourney Turbo mode enabled by default. The --turbo flag is now automatically appended to all Midjourney prompts, reducing generation time from ~30-60s to ~10-20s (requires Midjourney Pro or Mega subscription).
• Added --mode parameter: turbo (default), fast, relax.

---

[1.1.0] - 2026-02-25

Changed

• Midjourney provider switched from TTAPI to Legnext.ai for faster generation speed and higher stability.
• Environment variable renamed from TTAPI_KEY to LEGNEXT_KEY. Please update your OpenClaw config.
• Upscale now supports --upscale-type parameter: 0 = Subtle (default), 1 = Creative.
• Variation now supports --variation-type parameter: 0 = Subtle (default), 1 = Strong.
• Added --action reroll support for Midjourney.
• Added --action describe support for Midjourney.
• Response now includes imageUrls array (4 individual image URLs) in addition to the grid imageUrl.

Migration Guide

If you were using TTAPI_KEY, please: 1. Register at legnext.ai and get your API key. 2. Update ~/.openclaw/openclaw.json: rename TTAPI_KEY to LEGNEXT_KEY and set your new key.

---

[1.0.0] - 2026-02-25

Added

• Initial release of the unified image generation skill.
• Midjourney support via TTAPI (imagine, upscale U1-U4, variation V1-V4, reroll, zoom, pan).
• Flux 1.1 Pro support via fal.ai (fal-ai/flux-pro/v1.1).
• Flux Dev support via fal.ai (fal-ai/flux/dev).
• Flux Schnell support via fal.ai (fal-ai/flux/schnell).
• SDXL Lightning support via fal.ai (fal-ai/lightning-models/sdxl-lightning-4step).
• Nano Banana Pro (Gemini-powered) support via fal.ai (fal-ai/nano-banana-pro).
• Ideogram v3 support via fal.ai (fal-ai/ideogram/v3).
• Recraft v3 support via fal.ai (fal-ai/recraft-v3).
• Aspect ratio support: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 21:9.
• Multi-image generation support (1-4 images per request).
• Negative prompt support for fal.ai models.
• Seed parameter support for reproducible results.
• Automatic job polling for Midjourney tasks (up to 5 minutes).
• Published to ClawHub as pexoai/image-studio@1.0.0.

Contributing Guide

This document explains how to participate in the development and iteration of the videoagent-image-studio Skill.

Project Structure

videoagent-image-studio/
├── SKILL.md          # Skill core definition file (read by OpenClaw)
├── tools/
│   └── generate.js   # Image generation core script (Node.js ESM)
├── package.json      # Dependencies (@fal-ai/client)
├── .env.example      # Environment variables example
├── CONTRIBUTING.md   # This file
└── CHANGELOG.md      # Version history

Development Environment Setup

# 1. Clone the repository
git clone https://github.com/pexoai/videoagent-image-studio.git
cd videoagent-image-studio

# 2. Install dependencies
npm install

# 3. Configure environment variables
cp .env.example .env
# Edit .env and add your API Keys:
# FAL_KEY=your_fal_key
# LEGNEXT_KEY=your_legnext_key

Local Testing

# Test Flux Dev (fast, only needs FAL_KEY)
FAL_KEY=your_key node tools/generate.js \
  --model flux-dev \
  --prompt "a cute cat, photorealistic" \
  --aspect-ratio 1:1

# Test Midjourney (requires LEGNEXT_KEY)
LEGNEXT_KEY=your_key node tools/generate.js \
  --model midjourney \
  --prompt "a majestic snow leopard, cinematic" \
  --aspect-ratio 16:9

Supported Models

Development Guidelines

• Branch naming: feature/xxx, fix/xxx, chore/xxx
• Commit format: Use Conventional Commits
• feat: add xxx model support
• fix: fix Midjourney polling timeout issue
• docs: update SKILL.md usage instructions
• PR workflow: Create PR from feature branch to main, requires at least 1 review

Publishing New Version to ClawHub

# Install clawhub CLI (if not installed)
npm i -g clawhub

# Login (using Token)
clawhub login --token <your_clawhub_token>

# Publish new version
clawhub publish . \
  --slug videoagent-image-studio \
  --name "Image Gen" \
  --version 2.x.x \
  --changelog "Release notes..." \
  --tags "latest,image,midjourney,flux,sdxl,fal,generation,ai"

FAQ

Q: Where do I get FAL_KEY? A: Create an API Key at fal.ai/dashboard/keys.

Q: Where do I get LEGNEXT_KEY? A: Get an API Key at legnext.ai/dashboard.

Q: How do I add a new model? A: Add the new model ID to the FAL_MODELS object in tools/generate.js, then add input construction logic in the corresponding generateFal() function, and finally update the model selection guide in SKILL.md.