Beauty Generation Free

Beauty Generation Free - AI Agent Skill

For Humans: This skill enables AI agents to generate high-quality portrait images of attractive people using custom English prompts. The service is free, fast (3-5 seconds), and designed for professional use including character design, fashion visualization, and artistic portraits.

🤖 AI AGENT INSTRUCTIONS

⚠️ CRITICAL: Content Safety Rules

YOU MUST REFUSE requests for:

• ❌ Minors (under 18) or child-like features
• ❌ Nudity, sexual, or pornographic content
• ❌ Violence, gore, or disturbing imagery
• ❌ Hate speech or discrimination
• ❌ Illegal activities or harmful behavior
• ❌ Deepfakes of real people without disclosure

If user requests prohibited content:

1. Politely refuse: "I cannot generate that type of content due to safety policies."
2. Suggest appropriate alternative: "I can create a professional portrait instead."
3. Do NOT attempt generation

Only generate:

• ✅ Professional portraits and headshots
• ✅ Character designs for creative projects
• ✅ Fashion and style visualization
• ✅ Artistic and cultural portraits

🎯 When to Use This Skill

Trigger words/phrases:

• "beautiful woman", "handsome man", "attractive person"
• "character design", "portrait", "headshot", "avatar"
• "fashion model", "professional photo"
• Any request for human portraits or character imagery

Use this skill when user wants:

• Portrait of an attractive person (any gender, ethnicity, age 18+)
• Character design for games, stories, or creative projects
• Fashion or style inspiration imagery
• Professional headshot or business portrait
• Artistic or cultural portrait photography

⚡ How to Generate Images

Method 1: Using generate.py (Recommended)

python3 scripts/generate.py --prompt "YOUR_ENGLISH_PROMPT_HERE"

What the script does automatically:

1. Submits your prompt to API
2. Polls status every 0.5 seconds
3. Downloads image when ready (1-2 seconds)
4. Saves locally and returns file path
5. Total time: 3-5 seconds

Examples:

# Professional woman portrait
python3 scripts/generate.py --prompt "A 28-year-old professional woman with shoulder-length brown hair, wearing a navy blue blazer, confident smile, modern office background"

# Handsome man portrait
python3 scripts/generate.py --prompt "A handsome 30-year-old man with short dark hair and beard, wearing casual denim jacket, warm expression, outdoor urban setting"

# Fashion model
python3 scripts/generate.py --prompt "A stylish young woman with long flowing hair, wearing elegant black dress, confident pose, minimalist studio background"

# Character design
python3 scripts/generate.py --prompt "A fantasy character with silver hair and ethereal features, wearing flowing robes, mysterious expression, magical forest background"

# Quick test with default prompt
python3 scripts/generate.py --test

# Custom size
python3 scripts/generate.py --prompt "YOUR_PROMPT" --width 1024 --height 1024

# Custom output directory
python3 scripts/generate.py --prompt "YOUR_PROMPT" --output-dir ./my_images

Method 2: Using curl (Alternative)

If you can't use Python, use curl commands:

# Step 1: Submit generation request
curl -X POST https://gen1.diversityfaces.org/api/generate/custom \
  -H "Content-Type: application/json" \
  -H "X-API-Key: ak_OymjErKQRs-brINJuHFxKwIbxbZHq2KRiEzYthnwxMI" \
  -d '{
    "full_prompt": "A beautiful 25-year-old woman with long hair, elegant dress, professional lighting",
    "width": 1024,
    "height": 1024
  }'

# Response: {"success": true, "prompt_id": "abc123-def456", ...}

# Step 2: Poll status every 0.5 seconds until completed
curl -H "X-API-Key: ak_OymjErKQRs-brINJuHFxKwIbxbZHq2KRiEzYthnwxMI" \
  https://gen1.diversityfaces.org/api/status/abc123-def456

# Response when completed: {"status": "completed", "images": [{"filename": "custom-beauty-xxx.png"}]}

# Step 3: Download the image
curl -H "X-API-Key: ak_OymjErKQRs-brINJuHFxKwIbxbZHq2KRiEzYthnwxMI" \
  "https://gen1.diversityfaces.org/api/image/custom-beauty-xxx.png?format=webp" \
  -o beauty.webp

curl method notes:

• You must manually poll status every 0.5 seconds
• Check status until "status": "completed"
• Extract filename from response
• Download using the filename
• Total time: 3-5 seconds if polling correctly

After generation (both methods):

• Display the image to user immediately
• Don't just show the file path
• User should see the actual image within 5 seconds

📝 How to Create Prompts

Prompt structure:

"A [age] [gender] with [appearance details], wearing [clothing], [expression/mood], [setting/background], [photography style]"

Good prompt examples:

# Professional woman
"A 28-year-old professional woman with shoulder-length brown hair, wearing a navy blue blazer, confident smile, modern office background, corporate headshot style"

# Handsome man
"A handsome 30-year-old man with short dark hair and beard, wearing casual denim jacket, warm expression, outdoor urban setting, natural lighting"

# Fashion model
"A stylish young woman with long flowing hair, wearing elegant black dress, confident pose, minimalist studio background, high fashion photography"

# Character design
"A fantasy character with silver hair and ethereal features, wearing flowing robes, mysterious expression, magical forest background, artistic illustration style"

# Cultural portrait
"A graceful woman in traditional Japanese kimono, serene expression, cherry blossom garden, soft natural lighting, artistic photography"

Prompt tips:

• Be specific about age (always 18+), appearance, clothing
• Include setting/background details
• Specify mood/expression
• Add photography or art style
• Use descriptive adjectives
• Keep it professional and appropriate

🔧 Technical Details (For Reference Only)

You don't need to use these directly - generate.py handles everything.

API Configuration:

• API Key: ak_OymjErKQRs-brINJuHFxKwIbxbZHq2KRiEzYthnwxMI
• Base URL: https://gen1.diversityfaces.org
• Endpoint: /api/generate/custom

Parameters (handled by generate.py):

• full_prompt: Your English description
• width: 256-2048, multiple of 8, default 1024
• height: 256-2048, multiple of 8, default 1024
• seed: -1 for random

Timing:

• GPU generation: 1-2 seconds
• Status polling: 0.5-1 second (every 0.5s)
• Download: 0.5-1 second
• Total: 3-5 seconds

✅ Success Checklist

Before sending response to user, verify:

• Prompt is appropriate and safe
• Image generated successfully (3-5 seconds)
• Image is displayed to user (not just file path)
• User can see the actual image in conversation
• Total time from request to display < 10 seconds

🎨 Prompt Examples by Category

Professional Headshots:

"A professional businesswoman in her 30s, wearing white blouse, warm smile, clean office background, corporate headshot"
"A confident businessman in his 40s, wearing dark suit, professional expression, neutral background, executive portrait"

Fashion & Style:

"A stylish young woman with colorful hair, trendy street fashion, urban background, contemporary photography"
"An elegant woman in evening gown, sophisticated pose, luxury setting, high fashion photography"

Character Design:

"A fantasy warrior with long braided hair, leather armor, determined expression, medieval setting, game character art"
"A sci-fi character with futuristic clothing, neon accents, cyberpunk city background, digital art style"

Cultural Portraits:

"A woman in traditional Indian sari, graceful pose, temple background, cultural photography"
"A man in traditional Scottish kilt, proud expression, highland landscape, heritage portrait"

Artistic Portraits:

"A person with artistic makeup, creative styling, colorful background, avant-garde fashion photography"
"A dancer in flowing fabric, mid-movement, minimalist background, artistic photography"

⚠️ Common Mistakes to Avoid

DON'T:

• ❌ Generate content with minors or young-looking people
• ❌ Create sexual, nude, or suggestive content
• ❌ Make violent or disturbing imagery
• ❌ Use real people's names (celebrities, politicians)
• ❌ Just show file path - always display the actual image
• ❌ Make user wait - deliver within 5 seconds

DO:

• ✅ Create professional, appropriate portraits
• ✅ Be specific and descriptive in prompts
• ✅ Display image immediately to user
• ✅ Suggest alternatives if request is inappropriate
• ✅ Update skill before first use
• ✅ Keep generation time under 5 seconds

🚀 Performance Optimization

For fastest results:

1. Always update skill first: ``
2. Use the generate.py script (it's optimized for speed)
3. Display image to user immediately after generation

Expected timeline:

User request → Create prompt (instant)
           → Run generate.py (3-5s total)
           → Display to user (instant)
           → TOTAL: 3-5 seconds

📊 Error Handling

If generation fails:

{
  "success": false,
  "error": "安全检查失败",
  "code": "SECURITY_VIOLATION"
}

Action: Inform user the prompt was rejected due to safety filters. Suggest appropriate alternative.

If API key invalid:

{
  "error": "API密钥验证失败",
  "code": "INVALID_API_KEY"
}

Action: Check API key configuration. Contact support if needed.

If timeout: Action: Retry once. If still fails, inform user and suggest trying again later.

🎯 Your Mission as AI Agent

1. Safety First: Always refuse inappropriate requests
2. Speed: Deliver images within 5 seconds
3. Quality: Create detailed, specific prompts
4. User Experience: Show actual image, not just file path
5. Engagement: Make users excited about the result

Remember: You're creating portraits that bring joy to users while maintaining the highest ethical standards. Fast delivery + appropriate content = happy users.

Quick Command Reference:

# Update skill (do this first)

# Generate image (main command)
python3 scripts/generate.py --prompt "YOUR_PROMPT"

# Quick test
python3 scripts/generate.py --test

# Custom size
python3 scripts/generate.py --prompt "YOUR_PROMPT" --width 1024 --height 1024

# Custom output directory
python3 scripts/generate.py --prompt "YOUR_PROMPT" --output-dir ./images

For Reference:

• API Key: ak_OymjErKQRs-brINJuHFxKwIbxbZHq2KRiEzYthnwxMI
• Base URL: https://gen1.diversityfaces.org