API Integration

# Clone the skill repository
git clone https://github.com/xianyu110/awesome-openclaw-tutorial.git

# Copy to OpenClaw's skills directory
cp -r expert-skills-hub/skills/bananapro-image-gen ~/.openclaw/skills/

# Install dependencies
cd ~/.openclaw/skills/bananapro-image-gen
pip3 install -r requirements.txt

npx skills add https://github.com/xianyu110/awesome-openclaw-tutorial --skill bananapro-image-gen

ls ~/.openclaw/skills/bananapro-image-gen
# You should see: SKILL.md, README.md, scripts/, test.sh and other files

nano ~/.openclaw/openclaw.json

{
  "api": {
    "gemini": {
      "apiKey": "your-api-key-here",
      "baseUrl": "https://apipro.maynor1024.live",
      "model": "gemini-3-pro-image-preview"
    }
  }
}

cd ~/.openclaw/skills/bananapro-image-gen
bash test.sh

You: Help me design a logo with the theme "Super Individual", requirements:
- Simple and modern
- Use blue and orange
- Include elements of a person and AI
- Suitable for social media avatars

You: Generate a whiteboard image, handwritten font style, with the content:

Title: OpenClaw Core Advantages

4 Major Advantages:
1. Local Deployment - Privacy Protection
2. File Access - Smart Search  
3. Skills Extension - Infinite Possibilities
4. Multi-platform Support - Anytime, Anywhere

Use handwritten font, add arrows, boxes, and other hand-drawn elements

You: Generate an Xiaohongshu illustration with the theme "Super Individual in the AI Era", requirements:
- 9:16 vertical format
- Warm color scheme
- Include text: One Person + AI = Infinite Possibilities
- Minimalist style

cd ~/.openclaw/skills/bananapro-image-gen

# Generate image (OpenAI format)
python3 scripts/generate_image.py \
    --prompt "A minimalist logo design, theme is AI assistant" \
    --filename output.png \
    --api-format openai

# Generate high-definition image (Gemini format)
python3 scripts/generate_image.py \
    --prompt "A whiteboard diagram, handwritten style" \
    --filename whiteboard.png \
    --api-format gemini \
    --resolution 2K

# Create batch generation script
cat > batch_generate.sh << 'EOF'
#!/bin/bash

# Generate 5 images with different themes
prompts=(
    "Logo design: Super Individual"
    "Whiteboard diagram: AI workflow"
    "Poster: 10x efficiency improvement"
    "Illustration: Knowledge management system"
    "Icon: Automation tool"
)

for i in "${!prompts[@]}"; do
    echo "Generating image $((i+1))..."
    python3 scripts/generate_image.py \
        --prompt "${prompts[$i]}" \
        --filename "output_$((i+1)).png"
done

echo "Batch generation complete!"
EOF

chmod +x batch_generate.sh
./batch_generate.sh

❌ Bad: Generate an image
✅ Good: Generate a whiteboard image, handwritten font style, simple and modern

❌ Bad: Draw a logo
✅ Good: Design a logo, circular, blue gradient, with an abstract brain pattern in the center,
      surrounded by flowing data lines, with a modern tech feel

❌ Bad: Draw an architecture diagram
✅ Good: Draw a system architecture diagram, including:
      - User client (phone icon)
      - Gateway (cloud icon)
      - Database (cylinder icon)
      - Connect components with arrows
      - Add text labels

✅ Good: Generate an image, with a style similar to:
      - Hand-drawn whiteboard style
      - Flat design
      - Watercolor style
      - Pixel art style

# Logo Design Template
"Design a {theme} logo, requirements:
- Shape: [Circular/Square/Abstract]
- Color: [Main color tone]
- Elements: [Core elements]
- Style: [Modern/Retro/Minimalist]
- Usage: [Social media/Website/Business card]"

# Whiteboard Diagram Template
"Generate a whiteboard image, handwritten font style, with the content:
Title: [Title text]
Key points:
1. [Point 1]
2. [Point 2]
3. [Point 3]
Use handwritten font, add arrows, boxes, underlines, and other hand-drawn elements"

# Social Media Illustration Template
"Generate a [platform] illustration with the theme [theme], requirements:
- Size: [9:16/16:9/1:1]
- Color scheme: [Warm/Cool/Contrast]
- Text: [Core copy]
- Style: [Minimalist/Rich/Creative]"

# Architecture Diagram Template
"Generate an architecture diagram for [system name], including:
Components:
- [Component 1]: [Description]
- [Component 2]: [Description]
- [Component 3]: [Description]
Connection relationships:
- [Component 1] → [Component 2]
- [Component 2] → [Component 3]
Use boxes, arrows, icons, and other elements to clearly display the system structure"

# Use the provided test script
cd ~/.openclaw/skills/bananapro-image-gen
bash test_chapters.sh

# create_social_images.py
import subprocess
import json

# List of article titles
articles = [
    "Super Individual in the AI Era",
    "The Secret to 10x Efficiency Improvement",
    "One Person's Infinite Possibilities",
    # ... more titles
]

for i, title in enumerate(articles, 1):
    prompt = f"""Generate an Xiaohongshu illustration with the theme "{title}", requirements:
    - 9:16 vertical format
    - Warm color scheme
    - Include title text
    - Minimalist modern style
    """
    
    subprocess.run([
        "python3", "scripts/generate_image.py",
        "--prompt", prompt,
        "--filename", f"social_{i}.png",
        "--resolution", "2K"
    ])
    
    print(f"✅ Generating the {i}th illustration")

print("🎉 Batch generation complete!")

# Create Logo generation script
cat > generate_logos.sh << 'EOF'
#!/bin/bash

# Project list
projects=(
    "Super Individual:Blue:Modern Tech"
    "Knowledge Management:Green:Minimalist"
    "Efficiency Tool:Orange:Vibrant"
)

for project in "${projects[@]}"; do
    IFS=':' read -r name color style <<< "$project"
    
    prompt="Design a logo for ${name}, requirements:
    - Main color tone: ${color}
    - Style: ${style}
    - Simple and modern
    - Suitable for social media"
    
    python3 scripts/generate_image.py \
        --prompt "$prompt" \
        --filename "logo_${name}.png"
    
    echo "✅ Generating ${name} Logo"
done
EOF

chmod +x generate_logos.sh
./generate_logos.sh

# Convert to JPG
convert output.png output.jpg

# Convert to WebP (smaller file)
convert output.png output.webp

# In OpenClaw's automation script
from skills.nano_banana_image_gen import generate_image

# Generate image
image_path = generate_image(
    prompt="Your prompt",
    filename="output.png",
    resolution="1K"
)

print(f"Image generated: {image_path}")

# OpenAI format (good compatibility)
python3 scripts/generate_image.py \
    --api-format openai \
    --prompt "Your prompt"

# Gemini format (more powerful features)
python3 scripts/generate_image.py \
    --api-format gemini \
    --prompt "Your prompt" \
    --resolution 2K

from PIL import Image, ImageDraw, ImageFont

# Open the generated image
img = Image.open("output.png")

# Add watermark
draw = ImageDraw.Draw(img)
font = ImageFont.truetype("Arial.ttf", 36)
draw.text((10, 10), "© 2026", fill="white", font=font)

# Resize
img = img.resize((800, 600))

# Save
img.save("output_processed.png")

# image_templates.py
TEMPLATES = {
    "logo": """Design a {theme} logo, requirements:
    - Shape: {shape}
    - Color: {color}
    - Style: {style}
    - Usage: {usage}""",
    
    "whiteboard": """Generate a whiteboard image, handwritten font style, with the content:
    Title: {title}
    {content}
    Use handwritten font, add arrows, boxes, and other hand-drawn elements""",
    
    "social": """Generate a {platform} illustration with the theme {theme}, requirements:
    - Size: {ratio}
    - Color scheme: {color_scheme}
    - Text: {text}
    - Style: {style}"""
}

def generate_from_template(template_name, **kwargs):
    prompt = TEMPLATES[template_name].format(**kwargs)
    # Call the generation function
    return generate_image(prompt)

# Use template
generate_from_template(
    "logo",
    theme="AI Assistant",
    shape="Circular",
    color="Blue Gradient",
    style="Modern Tech",
    usage="Social Media"
)

# Install Notion SDK
npm install @notionhq/client

# Or use Python SDK
pip install notion-client

# Create Skill directory
mkdir -p ~/.openclaw/skills/my-notion-skill
cd ~/.openclaw/skills/my-notion-skill

# Create skill.json
cat > skill.json << 'EOF'
{
  "name": "my-notion-skill",
  "version": "1.0.0",
  "description": "Custom Notion integration",
  "main": "index.js"
}
EOF

# Install dependencies
npm init -y
npm install @notionhq/client

{
  "mcpServers": {
    "notion": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-notion"],
      "env": {
        "NOTION_API_KEY": "your-api-key"
      }
    }
  }
}

https://www.notion.so/workspace/DatabaseName-1234567890abcdef1234567890abcdef?v=...
                                                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                                                 This part is the Database ID

1234567890abcdef1234567890abcdef

# Use curl to query all accessible databases
curl -X POST https://api.notion.com/v1/search \
  -H "Authorization: Bearer secret_xxxxxxxxxxxxxxxxxxxxxxxxxxxxx" \
  -H "Notion-Version: 2022-06-28" \
  -H "Content-Type: application/json" \
  -d '{
    "filter": {
      "property": "object",
      "value": "database"
    }
  }'

# Add to OpenClaw configuration
nano ~/.openclaw/openclaw.json

{
  "api": {
    "notion": {
      "apiKey": "secret_xxxxxxxxxxxxxxxxxxxxxxxxxxxxx",
      "version": "2022-06-28"
    }
  },
  "notion": {
    "databases": {
      "notes": "1234567890abcdef1234567890abcdef",
      "tasks": "abcdef1234567890abcdef1234567890",
      "projects": "567890abcdef1234567890abcdef1234"
    },
    "defaultDatabase": "notes"
  }
}

# Set variables
NOTION_API_KEY="secret_xxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
NOTION_VERSION="2022-06-28"

# Test API connection
curl -X GET https://api.notion.com/v1/users/me \
  -H "Authorization: Bearer $NOTION_API_KEY" \
  -H "Notion-Version: $NOTION_VERSION"

# If successful, your user information will be returned

# Query specific database
DATABASE_ID="1234567890abcdef1234567890abcdef"

curl -X POST "https://api.notion.com/v1/databases/$DATABASE_ID/query" \
  -H "Authorization: Bearer $NOTION_API_KEY" \
  -H "Notion-Version: $NOTION_VERSION" \
  -H "Content-Type: application/json" \
  -d '{}'

# If successful, entries from the database will be returned

# Create a test page in the database
curl -X POST https://api.notion.com/v1/pages \
  -H "Authorization: Bearer $NOTION_API_KEY" \
  -H "Notion-Version: $NOTION_VERSION" \
  -H "Content-Type: application/json" \
  -d '{
    "parent": {
      "database_id": "'$DATABASE_ID'"
    },
    "properties": {
      "Name": {
        "title": [
          {
            "text": {
              "content": "OpenClaw Test Page"
            }
          }
        ]
      }
    }
  }'

# If successful, information about the newly created page will be returned

# If Notion Skill is installed
openclaw notion test

# Or test directly in OpenClaw
You: Test Notion connection

OpenClaw: Testing Notion API connection...

✅ API connection successful!

📊 Connection Information:
• API Version: 2022-06-28
• Username: Your Name
• Workspace: Your Workspace
• Accessible Pages: 15
• Accessible Databases: 3

📁 Available Databases:
1. Notes Database (notes)
2. Tasks Database (tasks)
3. Projects Database (projects)

✅ Notion integration configured successfully!

# Check API Key format
echo $NOTION_API_KEY
# Should be: secret_xxxxxxxxxxxxxxxxxxxxxxxxxxxxx

# Re-copy API Key
# Visit https://www.notion.so/my-integrations
# Re-copy the key and update the configuration

# Use API to search all databases
curl -X POST https://api.notion.com/v1/search \
  -H "Authorization: Bearer $NOTION_API_KEY" \
  -H "Notion-Version: $NOTION_VERSION" \
  -H "Content-Type: application/json" \
  -d '{
    "filter": {
      "property": "object",
      "value": "database"
    }
  }' | jq '.results[] | {id: .id, title: .title}'

# Set configuration file permissions
chmod 600 ~/.openclaw/config/notion.json

# Do not commit API Key to Git
echo "*.json" >> ~/.openclaw/.gitignore
echo "openclaw.json" >> ~/.gitignore

# Set in ~/.bashrc or ~/.zshrc
export NOTION_API_KEY="secret_xxxxxxxxxxxxxxxxxxxxxxxxxxxxx"

# Use in scripts
NOTION_API_KEY="${NOTION_API_KEY}"

# Change API Key every 3-6 months
# 1. Create a new Integration in Notion
# 2. Update configuration file
# 3. Delete the old Integration

{
  "api": {
    "notion": {
      "apiKey": "secret_xxxxxxxxxxxxxxxxxxxxxxxxxxxxx",
      "version": "2022-06-28",
      "timeout": 30000,
      "retries": 3
    }
  },
  "notion": {
    "databases": {
      "notes": "1234567890abcdef1234567890abcdef",
      "tasks": "abcdef1234567890abcdef1234567890",
      "projects": "567890abcdef1234567890abcdef1234",
      "meetings": "890abcdef1234567890abcdef1234567"
    },
    "defaultDatabase": "notes",
    "autoSync": true,
    "syncInterval": 300,
    "templates": {
      "note": {
        "properties": {
          "Tags": ["Study Notes"],
          "Status": "Draft"
        }
      },
      "task": {
        "properties": {
          "Status": "To Do",
          "Priority": "Medium"
        }
      }
    }
  }
}

# Edit OpenClaw configuration
nano ~/.openclaw/openclaw.json

{
  "api": {
    "notion": {
      "apiKey": "secret_xxxxxxxxxxxxx",
      "version": "2022-06-28"
    }
  },
  "notion": {
    "defaultDatabase": "your-database-id",
    "autoSync": true,
    "syncInterval": 300
  }
}

You: Save this article to Notion

Title: OpenClaw Usage Experience
Content:
Today I learned about OpenClaw's API integration feature,
Key takeaways include:
1. Learned how to configure Gemini API
2. Successfully generated the first AI image
3. Understood the Skills extension mechanism

OpenClaw: Okay, saving to Notion...

📝 [Create Notion Page]

✅ Page created successfully!

📄 Page Information:
• Title: OpenClaw Usage Experience
• Creation Time: 2026-02-12 10:30
• Page ID: abc123...
• Link: https://notion.so/abc123

📊 Content Statistics:
• Word Count: 120 words
• Paragraphs: 4
• List Items: 3

Automatically added:
• Tags: #StudyNotes #OpenClaw
• Category: Technical Documentation
• Creation Date: 2026-02-12

🔗 Open in Notion:
https://notion.so/abc123

You: Import all Markdown files in this folder to Notion

Folder: ~/Documents/notes/

OpenClaw: Okay, scanning files...

📁 [Batch Import to Notion]

📊 Scan Results:
• Folder: ~/Documents/notes/
• Number of Files: 15
• Total Size: 2.5MB
• Estimated Time: 2 minutes

⏳ Import Progress:

[████████████████████] 100% (15/15)

✅ Import complete!

📊 Import Statistics:
• Success: 15
• Failed: 0
• Total Time: 1 minute 45 seconds

📁 Notion Structure:
Notes Import/
├── Study Notes/
│   ├── Study Notes.md
│   └── Technical Documentation.md
├── Project Management/
│   ├── Project Planning.md
│   └── Meeting Minutes.md
└── Product Design/
    └── Product Requirements.md

🔗 View in Notion:
https://notion.so/workspace/notes-import

# Create meeting minutes template
openclaw notion template create "meeting" \
  --title "Meeting Minutes-{date}" \
  --properties "Topic,Time,Attendees,Content,Decisions,Tasks" \
  --database "meetings-db-id"

# Use template
You: Create today's meeting minutes using the meeting template

# Set auto-sync rules
openclaw config set notion.auto-rules '{
  "markdown-to-notion": {
    "watch": "~/Documents/notes/",
    "target": "notes-db-id",
    "trigger": "file-save"
  },
  "task-sync": {
    "watch": "~/Projects/tasks.md",
    "target": "tasks-db-id",
    "trigger": "file-change"
  }
}'

# Enable bi-directional synchronization
openclaw config set notion.bidirectional true

# Notion updates will automatically sync to local
# Local updates will automatically sync to Notion

# Test API connection
openclaw notion test

# View error logs
openclaw logs --filter notion

# Batch import (automatic batching)
openclaw notion import ~/notes/ \
  --batch-size 10 \
  --delay 1000

# Paginated query
openclaw notion query \
  --database "db-id" \
  --page-size 100

# Configure deduplication rules
openclaw config set notion.dedup '{
  "enabled": true,
  "field": "title",
  "action": "skip"
}'

# Install
npx clawhub@latest install video-agent

# Features
• Generate AI avatar videos using HeyGen API
• Supports multiple languages and voice tones
• Professional-grade video quality
• Suitable for tutorials and marketing videos

# Install
npx clawhub@latest install sora-video-gen

# Features
• Uses OpenAI Sora API
• Text-to-video generation
• High-quality video output
• Creative video production

# Install
npx clawhub@latest install veo3-video-gen

# Features
• Uses Google Veo 3.x
• Generate and stitch short videos
• Supports various styles
• Suitable for short video creation

# Install
npx clawhub@latest install tube-cog

# Features
• YouTube content creation
• Powered by CellCog
• Complete video production workflow
• Automated publishing

# Install
npx clawhub@latest install video-cog

# Features
• Long-form AI video production
• Multi-agent collaboration
• Professional-grade video output
• Suitable for tutorials and courses

# Edit OpenClaw configuration
nano ~/.openclaw/openclaw.json

{
  "api": {
    "heygen": {
      "apiKey": "your-heygen-api-key",
      "baseUrl": "https://api.heygen.com/v1"
    }
  },
  "video": {
    "defaultAvatar": "avatar-id",
    "defaultVoice": "voice-id",
    "quality": "high",
    "format": "mp4"
  }
}

You: Generate a 30-second AI avatar video with the content:

Theme: OpenClaw Quick Start
Script:
Hello, I am OpenClaw AI Assistant.
Today, I will introduce how to quickly get started with OpenClaw.
Just 3 steps: Install, Configure, Use.
Let's begin!

Requirements:
- Avatar: Professional male
- Voice Tone: Gentle
- Background: Tech-inspired
- Subtitles: Chinese

OpenClaw: Okay, generating video...

🎬 [AI Video Generation] OpenClaw Quick Start
📊 Video Information:
• Duration: 30 seconds
• Resolution: 1080p
• Format: MP4
• Size: Approx. 15MB

⏳ Generation Progress:

1. Script Analysis [████████████████████] 100%
2. Avatar Rendering [████████████████████] 100%
3. Voice Synthesis [████████████████████] 100%
4. Video Composition [████████████████████] 100%
5. Subtitle Addition [████████████████████] 100%

✅ Video generation complete!

📹 Video Preview:
[Video Thumbnail]

📊 Video Details:
• Avatar: Professional Male (ID: avatar-001)
• Voice Tone: Gentle Male Voice (ID: voice-cn-male-01)
• Background: Tech Blue Gradient
• Subtitles: Chinese, White, Bottom Centered

💰 Cost:
• Generation Fee: $1.00
• Storage Fee: $0.01
• Total: $1.01

📁 Files Saved:
• Video File: openclaw_intro.mp4
• Subtitle File: openclaw_intro.srt
• Script File: openclaw_intro.txt

🔗 Online Preview:
https://video.openclaw.com/abc123

You: Batch generate an OpenClaw tutorial video series, 5 episodes in total.

Episode 1: What is OpenClaw
Episode 2: How to Install OpenClaw
Episode 3: Basic Configuration Tutorial
Episode 4: Using Skills Extensions
Episode 5: Practical Case Sharing

OpenClaw: Okay, batch generating now...

🎬 [Batch Video Generation] OpenClaw Tutorial Series

📊 Task Information:
• Number of Videos: 5 episodes
• Duration per Episode: 3-5 minutes
• Total Duration: Approx. 20 minutes
• Estimated Time: 15 minutes

⏳ Generation Progress:

Episode 1: What is OpenClaw
[████████████████████] 100% ✅

Episode 2: How to Install OpenClaw
[████████████████████] 100% ✅

Episode 3: Basic Configuration Tutorial
[████████████████████] 100% ✅

Episode 4: Using Skills Extensions
[████████████████████] 100% ✅

Episode 5: Practical Case Sharing
[████████████████████] 100% ✅

✅ All generated!

📊 Generation Statistics:
• Success: 5 episodes
• Failure: 0 episodes
• Total Time Spent: 12 minutes 30 seconds
• Total Size: 150MB

📁 File List:
1. ✅ 01_What_is_OpenClaw.mp4 (30MB)
2. ✅ 02_How_to_Install_OpenClaw.mp4 (28MB)
3. ✅ 03_Basic_Configuration_Tutorial.mp4 (32MB)
4. ✅ 04_Skills_Extension_Usage.mp4 (35MB)
5. ✅ 05_Practical_Case_Sharing.mp4 (25MB)

💰 Total Cost:
• Generation Fee: $15.00
• Storage Fee: $0.15
• Total: $15.15

📦 Packaged:
• OpenClaw Tutorial Series.zip (150MB)

🔗 Download Link:
https://video.openclaw.com/series/abc123

# Create an automation script
cat > daily_video.sh << 'EOF'
#!/bin/bash

# Get today's topic
topic=$(openclaw ask "Recommend 1 technical topic suitable for a short video")

# Generate script
script=$(openclaw ask "Generate a 60-second short video script for the topic '${topic}'")

# Generate video
openclaw video generate \
  --script "$script" \
  --duration 60 \
  --avatar "professional-male" \
  --voice "cn-male-01" \
  --output "daily_$(date +%Y%m%d).mp4"

echo "✅ Today's video generated: ${topic}"
EOF

chmod +x daily_video.sh

# Set up a scheduled task (every day at 9 AM)
openclaw schedule add "daily-video" \
  --time "09:00" \
  --command "./daily_video.sh"

You: Generate product introduction videos for OpenClaw, requiring 3 versions: Chinese, English, and Japanese.

Product Features:
1. Local deployment, privacy protection
2. File access, intelligent search
3. Skills extensions, infinite possibilities
4. Multi-platform support, anytime, anywhere

OpenClaw: Okay, generating multi-language versions now...

🎬 [Multi-language Video Generation]

📊 Task Information:
• Language Versions: 3 (Chinese, English, Japanese)
• Video Duration: 2 minutes each
• Estimated Time: 10 minutes

⏳ Generation Progress:

Chinese Version:
[████████████████████] 100% ✅

English Version:
[████████████████████] 100% ✅

Japanese Version:
[████████████████████] 100% ✅

✅ All generated!

📁 File List:
1. ✅ OpenClaw_Intro_CN.mp4 (40MB)
2. ✅ OpenClaw_Intro_EN.mp4 (38MB)
3. ✅ OpenClaw_Intro_JP.mp4 (39MB)

💰 Total Cost: $6.00

🔗 Online Preview:
• Chinese: https://video.openclaw.com/cn/abc123
• English: https://video.openclaw.com/en/abc123
• Japanese: https://video.openclaw.com/jp/abc123

# Upload a custom avatar photo
openclaw video avatar upload \
  --photo "my_photo.jpg" \
  --name "my-avatar"

# Use a custom avatar
openclaw video generate \
  --avatar "my-avatar" \
  --script "Your script"

# Add intro and outro
openclaw video edit \
  --input "video.mp4" \
  --intro "intro.mp4" \
  --outro "outro.mp4" \
  --output "final.mp4"

# Add background music
openclaw video edit \
  --input "video.mp4" \
  --bgm "music.mp3" \
  --volume 0.3 \
  --output "with_music.mp4"

# Generate multiple videos in parallel
openclaw video batch \
  --scripts "scripts/*.txt" \
  --parallel 3 \
  --output "videos/"

# Install
npx clawhub@latest install elevenlabs

# Features
• Hyper-realistic voice quality
• Supports multiple languages and voice tones
• Rich emotional expression
• Suitable for professional narration

# Install
npx clawhub@latest install azure-tts

# Features
• Supports 140+ languages
• Neural network voices
• High-quality output
• Enterprise-grade stability

# Install
npx clawhub@latest install google-tts

# Features
• WaveNet technology
• Natural and fluent
• Multi-language support
• Cost-effective

# Install
npx clawhub@latest install openai-tts

# Features
• Uses OpenAI TTS API
• 6 high-quality voice tones
• Supports multiple languages
• Simple and easy to use

# Edit OpenClaw Configuration
nano ~/.openclaw/openclaw.json

{
  "api": {
    "elevenlabs": {
      "apiKey": "your-elevenlabs-api-key",
      "baseUrl": "https://api.elevenlabs.io/v1"
    }
  },
  "tts": {
    "defaultVoice": "voice-id",
    "defaultModel": "eleven_multilingual_v2",
    "quality": "high",
    "format": "mp3"
  }
}

You: Convert this text to speech:

OpenClaw is an open-source AI assistant framework.
It can access local files, operating system applications,
and extend various functions, truly making AI your assistant.

Requirements:
- Voice Tone: Professional Male
- Speech Rate: Normal
- Format: MP3
- Quality: High Definition

OpenClaw: Okay, converting now...

🔊 [Text-to-Speech]

📝 Text Information:
• Word Count: 52 characters
• Language: Chinese
• Estimated Duration: Approx. 15 seconds

⏳ Conversion Progress:

1. Text Analysis [████████████████████] 100%
2. Voice Synthesis [████████████████████] 100%
3. Audio Processing [████████████████████] 100%
4. Format Conversion [████████████████████] 100%

✅ Conversion complete!

🎵 Audio Information:
• Duration: 15 seconds
• Format: MP3
• Bitrate: 192kbps
• Size: 360KB

📊 Audio Details:
• Voice Tone: Professional Male (ID: voice-cn-male-01)
• Speech Rate: Normal (1.0x)
• Volume: Standard (0dB)
• Quality: High Definition

💰 Cost:
• Conversion Fee: $0.02
• Storage Fee: $0.001
• Total: $0.021

📁 Files Saved:
• Audio File: openclaw_intro.mp3
• Text File: openclaw_intro.txt

🔗 Online Preview:
https://audio.openclaw.com/abc123

You: Convert all articles in this folder to audio.

Folder: ~/Documents/articles/

OpenClaw: Okay, scanning files now...

🔊 [Batch Text-to-Speech]

📊 Scan Results:
• Folder: ~/Documents/articles/
• Number of Files: 10 articles
• Total Word Count: 25,000 characters
• Estimated Duration: Approx. 50 minutes
• Estimated Time: 5 minutes

⏳ Conversion Progress:

[████████████████████] 100% (10/10)

✅ All conversions complete!

📊 Conversion Statistics:
• Success: 10 articles
• Failure: 0 articles
• Total Time Spent: 4 minutes 30 seconds
• Total Duration: 48 minutes
• Total Size: 115MB

📁 File List:
1. ✅ Super_Individual_in_AI_Era.mp3 (12MB, 5 minutes)
2. ✅ Secret_to_10x_Efficiency.mp3 (10MB, 4 minutes)
3. ✅ Infinite_Possibilities_of_One_Person.mp3 (15MB, 6 minutes)
4. ✅ Building_a_Knowledge_Management_System.mp3 (11MB, 4.5 minutes)
5. ✅ Automated_Workflow_Design.mp3 (13MB, 5.5 minutes)
... (10 files in total)

💰 Total Cost:
• Conversion Fee: $7.50
• Storage Fee: $0.12
• Total: $7.62

📦 Packaged:
• Audiobook_Collection.zip (115MB)

🔗 Download Link:
https://audio.openclaw.com/batch/abc123

You: Convert this e-book into an audiobook.

File: The_Complete_Guide_to_OpenClaw.pdf
Requirements:
- Convert by chapter
- Each chapter as a separate file
- Add chapter title voiceovers
- Generate a playlist

OpenClaw: Okay, processing now...

📚 [Audiobook Production]

📖 E-book Information:
• Title: The Complete Guide to OpenClaw
• Number of Chapters: 16 chapters
• Total Word Count: 80,000 characters
• Estimated Duration: Approx. 3 hours

⏳ Processing Progress:

1. PDF Text Extraction [████████████████████] 100%
2. Chapter Segmentation [████████████████████] 100%
3. Voice Synthesis [████████████████████] 100%
4. Audio Processing [████████████████████] 100%
5. Playlist Generation [████████████████████] 100%

✅ Production complete!

📁 File Structure:
OpenClaw_Complete_Guide_Audio_Version/
├── 00_Title_and_Foreword.mp3 (5 minutes)
├── 01_Chapter_1_Introduction_to_OpenClaw.mp3 (10 minutes)
├── 02_Chapter_2_Quick_Start.mp3 (12 minutes)
├── 03_Chapter_3_Basic_Configuration.mp3 (15 minutes)
... (17 files in total)
├── playlist.m3u (Playlist)
└── metadata.json (Metadata)

📊 Production Statistics:
• Total Duration: 3 hours 5 minutes
• Total Size: 450MB
• Audio Format: MP3 192kbps
• Number of Chapters: 16 chapters + Foreword

💰 Total Cost: $24.00

📦 Packaged:
• OpenClaw_Complete_Guide_Audio_Version.zip (450MB)

🔗 Download Link:
https://audio.openclaw.com/audiobook/abc123

You: Generate voiceovers in Chinese, English, and Japanese for this video script.

Script:
OpenClaw is an open-source AI assistant framework.
It can access local files, operating system applications.
Truly making AI your super assistant.

OpenClaw: Okay, generating multi-language voiceovers now...

🎙️ [Multi-language Voiceover Generation]

📊 Task Information:
• Language Versions: 3 (Chinese, English, Japanese)
• Script Word Count: 52 characters
• Estimated Duration: Approx. 15 seconds each

⏳ Generation Progress:

Chinese Voiceover:
[████████████████████] 100% ✅

English Voiceover:
[████████████████████] 100% ✅

Japanese Voiceover:
[████████████████████] 100% ✅

✅ All generated!

📁 File List:
1. ✅ openclaw_intro_cn.mp3 (360KB, 15 seconds)
2. ✅ openclaw_intro_en.mp3 (380KB, 16 seconds)
3. ✅ openclaw_intro_jp.mp3 (370KB, 15 seconds)

📊 Voiceover Details:
• Chinese: Professional Male, normal speech rate
• English: American Male, normal speech rate
• Japanese: Standard Male, normal speech rate

💰 Total Cost: $0.06

🔗 Online Preview:
• Chinese: https://audio.openclaw.com/cn/abc123
• English: https://audio.openclaw.com/en/abc123
• Japanese: https://audio.openclaw.com/jp/abc123

# Clone your own voice (ElevenLabs)
openclaw tts voice clone \
  --samples "voice_samples/*.mp3" \
  --name "my-voice"

# Use a custom voice tone
openclaw tts generate \
  --text "Your text" \
  --voice "my-voice" \
  --output "output.mp3"

# Add emotional tags
openclaw tts generate \
  --text "This is truly amazing! [excited]" \
  --voice "voice-id" \
  --emotion "excited" \
  --output "excited.mp3"

# Adjust speech rate and pitch
openclaw tts generate \
  --text "Your text" \
  --voice "voice-id" \
  --speed 1.2 \
  --pitch 1.1 \
  --output "adjusted.mp3"

# Convert multiple files in parallel
openclaw tts batch \
  --input "texts/*.txt" \
  --voice "voice-id" \
  --parallel 5 \
  --output "audios/"

# Automatic segmentation for long texts
openclaw tts generate \
  --text-file "long_article.txt" \
  --auto-split true \
  --max-length 5000 \
  --output "output.mp3"

# Specify output format
openclaw tts generate \
  --text "Your text" \
  --format "wav" \
  --output "output.wav"

# Download and run the batch installation script
curl -O https://raw.githubusercontent.com/xianyu110/awesome-openclaw-tutorial/main/scripts/install_api_skills.sh
chmod +x install_api_skills.sh

# View usage help
./install_api_skills.sh --help

# Install Skills for specific scenarios
./install_api_skills.sh content    # Content creators
./install_api_skills.sh video      # Video creators
./install_api_skills.sh knowledge  # Knowledge workers
./install_api_skills.sh marketing  # Marketers
./install_api_skills.sh education  # Educators

# Install all Skills
./install_api_skills.sh all

# AI Image Generation Service Skills (verify existence first)
# Before use, it is recommended to search: npx clawhub@latest search <skill-name>
npx clawhub@latest install fal-ai
npx clawhub@latest install nvidia-image-gen
npx clawhub@latest install pollinations
npx clawhub@latest install venice-ai
npx clawhub@latest install recraft

# Notion Integration (ClawHub may not have a ready-made Skill currently)
# It is recommended to use the Notion API directly or create a custom Skill.

# Video Generation Service Skills
npx clawhub@latest install video-agent
npx clawhub@latest install sora-video-gen
npx clawhub@latest install veo3-video-gen
npx clawhub@latest install tube-cog
npx clawhub@latest install video-cog

# Voice Synthesis Service Skills
npx clawhub@latest install elevenlabs
npx clawhub@latest install azure-tts
npx clawhub@latest install google-tts
npx clawhub@latest install openai-tts

npx clawhub@latest install fal-ai
npx clawhub@latest install elevenlabs
# For Notion, direct API integration is recommended.

npx clawhub@latest install video-agent
npx clawhub@latest install tube-cog
npx clawhub@latest install elevenlabs

npx clawhub@latest install openai-tts
# For Notion, direct API integration is recommended.

npx clawhub@latest install fal-ai
npx clawhub@latest install video-agent
npx clawhub@latest install elevenlabs

npx clawhub@latest install fal-ai
npx clawhub@latest install video-cog
npx clawhub@latest install openai-tts

✅ Use domestic models (DeepSeek, Kimi)
✅ Batch process tasks
✅ Reuse generated results
✅ Set Token limits
✅ Choose appropriate quality levels
✅ Utilize free tiers
✅ Purchase in bulk for discounts

# Check network connection
ping github.com

# Use a domestic mirror
npm config set registry https://registry.npmmirror.com

# Retry installation
npx clawhub@latest install <skill-name>

# View detailed error
npx clawhub@latest install <skill-name> --verbose

# View Skills list
openclaw skills list

# Check Skills status
openclaw skills check

# View Skill detailed information
openclaw skills info <skill-name>

# Restart OpenClaw
openclaw restart

# Check API configuration
openclaw config get api

# Test API connection
openclaw api test <service-name>

# View error logs
openclaw logs --tail 50 --filter <service-name>

# Verify API key
openclaw api verify <service-name>

# Update a single Skill
npx clawhub@latest update <skill-name>

# Update all Skills
npx clawhub@latest update --all

# View updatable Skills
npx clawhub@latest list --outdated

# View update log
npx clawhub@latest changelog <skill-name>

# Uninstall a single Skill
npx clawhub@latest uninstall <skill-name>

# Uninstall multiple Skills
npx clawhub@latest uninstall <skill1> <skill2> <skill3>

# Check Skills dependencies
openclaw skills check

Week 1: Install basic Skills (notion, openai-tts)
      Familiarize with basic usage, establish workflow

Week 2: Add creative Skills (fal-ai, video-agent)
      Expand functionality based on needs

Week 3: Optimize configuration and cost
      Adjust parameters, improve efficiency

Week 4: Establish automated workflows
      Integrate all Skills, achieve automation

# Weekly update check
npx clawhub@latest list --outdated

# Periodically check Skills status
openclaw skills check

# Quarterly backup configuration
openclaw backup create

# Bi-annually evaluate usage
openclaw stats --period 6m

# Only load necessary Skills
openclaw config set skills.lazy-load true

# Set concurrency limit
openclaw config set skills.max-concurrent 3

# Enable caching
openclaw config set skills.cache.enabled true
openclaw config set skills.cache.ttl 3600

# Set timeout
openclaw config set skills.timeout 30000

# View API usage statistics
openclaw stats api --period 1m

# Set cost alerts
openclaw config set cost.alert.enabled true
openclaw config set cost.alert.threshold 100

# View cost report
openclaw cost report --period 1m

# Export cost data
openclaw cost export --format csv

# Create content creation profile
cat > ~/.openclaw/profiles/content-creation.json <<EOF
{
  "skills": ["fal-ai", "video-agent", "elevenlabs", "notion"],
  "description": "Content creation toolset"
}
EOF

# Create knowledge management profile
cat > ~/.openclaw/profiles/knowledge-management.json <<EOF
{
  "skills": ["notion", "notion-mcp", "openai-tts"],
  "description": "Knowledge management toolset"
}
EOF

# Create shortcut commands
openclaw alias create "画图" "Generate image using fal-ai"
openclaw alias create "记笔记" "Save to Notion"
openclaw alias create "配音" "Convert text to speech using elevenlabs"
openclaw alias create "做视频" "Generate video using video-agent"

# Use shortcut commands
You: 画图 a cute little lobster
You: 记笔记 Today I learned about API integration
You: 配音 this text
You: 做视频 product introduction script

# Create a batch installation script
cat > install_skills.sh << 'EOF'
#!/bin/bash

# Define Skills to install
skills=(
    "fal-ai"
    "notion"
    "elevenlabs"
    "video-agent"
)

echo "Starting batch installation of Skills..."

for skill in "${skills[@]}"; do
    echo "Installing: $skill"
    npx clawhub@latest install "$skill"
    
    if [ $? -eq 0 ]; then
        echo "✅ $skill installed successfully"
    else
        echo "❌ $skill installation failed"
    fi
    echo "---"
done

echo "Batch installation complete!"
EOF

chmod +x install_skills.sh
./install_skills.sh

# Batch generate images
openclaw batch run "fal-ai" \
  --input "prompts.txt" \
  --output "images/" \
  --parallel 3

# Batch convert audio
openclaw batch run "elevenlabs" \
  --input "texts/" \
  --output "audios/" \
  --parallel 5

# Batch import to Notion
openclaw batch run "notion" \
  --input "notes/" \
  --database "db-id" \
  --parallel 2

# Create an automated workflow
openclaw workflow create "daily-content" \
  --trigger "schedule:09:00" \
  --steps "
    1. Generate today's topic (ask)
    2. Generate illustration (fal-ai)
    3. Generate video (video-agent)
    4. Generate voiceover (elevenlabs)
    5. Save to Notion (notion)
  "

# Enable workflow
openclaw workflow enable "daily-content"

# View workflow status
openclaw workflow status "daily-content"

# View Skill dependencies
openclaw skills info <skill-name>

# Reinstall dependencies
cd ~/.openclaw/skills/<skill-name>
npm install

# Check configuration
openclaw config validate

# Fix permissions
chmod -R 755 ~/.openclaw/skills/

# Increase timeout duration
openclaw config set api.timeout 60000

# Use a proxy
openclaw config set api.proxy "http://proxy:port"

# Retry mechanism
openclaw config set api.retry.enabled true
openclaw config set api.retry.max 3

# Set daily limit
openclaw config set cost.daily-limit 10

# Lower quality settings
openclaw config set image.quality "standard"
openclaw config set video.quality "medium"

# Enable caching
openclaw config set cache.enabled true

# View cost details
openclaw cost detail --period 1d

✅ AI Drawing: From text to image
✅ Knowledge Management: Automated notes and databases
✅ Video Creation: From script to finished video
✅ Voice Synthesis: From text to sound

✅ Drawing: 30 minutes → 1 minute
✅ Notes: 5 minutes → 5 seconds
✅ Video: 2 hours → 10 minutes
✅ Voiceover: 1 hour → 2 minutes

✅ Drawing Cost: $50 → $0.1
✅ Notion: Free
✅ Video Cost: $200 → $2
✅ Voiceover Cost: $100 → $0.3

✅ Professional-grade output quality
✅ Multi-language support
✅ Batch processing capability
✅ Automated workflows