TaskAGI UGC Video Generation Workflow Setup Guide

What This Agent Does

This powerful workflow automates the creation of User-Generated Content (UGC) product advertisement videos directly from product images shared via Telegram. Simply send a product photo to your bot, and the system intelligently analyzes the product, generates a professional ad script, creates a polished video, and delivers it back to you—all without manual intervention.

Key benefits and time savings:

• 90% faster content creation: Transform product images into finished videos in minutes instead of hours
• Professional quality: Leverages AI-powered script writing and video generation for consistent, engaging ads
• Seamless workflow: Integrated Telegram interface means no context switching between platforms
• Scalable production: Process unlimited product images with consistent quality and branding
• Cost-effective: Automate repetitive creative tasks and reduce dependency on freelance creators

Target use cases:

• E-commerce businesses needing rapid product video ads
• Social media marketers managing multiple product launches
• Dropshipping operations requiring bulk UGC content
• Digital agencies producing ads for multiple clients
• Content creators building video libraries for TikTok, Instagram Reels, and YouTube Shorts

Who Is It For

This workflow is ideal for:

• E-commerce entrepreneurs who need to produce high-volume product ads quickly
• Marketing teams managing campaigns across multiple platforms and products
• Digital agencies delivering UGC content to clients at scale
• Social media managers constantly needing fresh video content
• Dropshippers launching new products weekly and needing promotional videos
• Content creators looking to automate repetitive production tasks

No video editing experience required—the AI handles all creative and technical aspects.

Required Integrations

Telegram

Why it's needed: Telegram serves as your primary interface for this workflow. It receives product images from users, triggers the entire automation pipeline, and delivers the finished videos back. This creates a seamless, user-friendly experience where creators simply upload images and receive videos.

Setup steps:

1. Create a Telegram Bot

   • Open Telegram and search for @BotFather
   • Send the command /newbot
   • Follow the prompts to name your bot (e.g., "UGC Video Generator")
   • Copy the API token provided (format: 123456789:ABCdefGHIjklmnoPQRstuvWXYZabcdefgh)

2. Obtain Your Bot Token

   • Save the token in a secure location
   • This is your primary credential for Telegram integration

3. Set Up Webhook in TaskAGI

   • In TaskAGI, navigate to Integrations → Telegram
   • Select Add New Integration
   • Paste your bot token in the API Token field
   • Click Authenticate
   • TaskAGI will generate a webhook URL

4. Configure Webhook in BotFather

   • Return to @BotFather in Telegram
   • Send /setwebhook
   • Provide the webhook URL from TaskAGI
   • Confirm the webhook is active

5. Test the Connection

   • Send a test message to your bot
   • Verify it appears in TaskAGI's activity logs

How to obtain API keys/credentials:

• Bot tokens are generated automatically by BotFather
• Keep your token confidential—treat it like a password
• You can regenerate tokens anytime via BotFather if compromised

Configuration in TaskAGI:

• Store the token in your Secrets Manager for security
• Reference it as {{secrets.telegram_token}} in node configurations
• Enable message logging for debugging

Google Gemini

Why it's needed: Google Gemini's advanced vision AI analyzes product images and generates compelling, professional UGC ad scripts. It understands product features, benefits, and emotional appeals—creating scripts optimized for conversion.

Setup steps:

1. Access Google Cloud Console

   • Visit console.cloud.google.com
   • Sign in with your Google account
   • Create a new project (name it "TaskAGI UGC Generator")

2. Enable the Gemini API

   • In the Cloud Console, go to APIs & Services → Library
   • Search for "Generative Language API"
   • Click the result and select Enable
   • Wait for activation (usually instant)

3. Create an API Key

   • Navigate to APIs & Services → Credentials
   • Click Create Credentials → API Key
   • Copy the generated key (format: AIzaSy...)
   • Restrict the key to Generative Language API only for security

4. Configure in TaskAGI

   • Go to Integrations → Google Gemini
   • Select Add New Integration
   • Paste your API key
   • Click Test Connection
   • Verify success message appears

5. Set Model Parameters

   • Model: gemini-2.5-flash (optimized for speed and quality)
   • Temperature: 0.7 (balanced creativity and consistency)
   • Max tokens: 1500 (sufficient for detailed scripts)

How to obtain API keys/credentials:

• API keys are generated in Google Cloud Console
• Free tier includes generous monthly quotas
• Monitor usage in APIs & Services → Quotas

Configuration in TaskAGI:

• Store API key in Secrets Manager as {{secrets.gemini_api_key}}
• Set up billing alerts to avoid unexpected charges
• Test with sample images before production use

Google Veo 3

Why it's needed: Google Veo 3 is a cutting-edge video generation AI that transforms scripts into polished, professional UGC-style videos. It handles cinematography, pacing, transitions, and visual effects—producing broadcast-quality content automatically.

Setup steps:

1. Access Google AI Studio

   • Visit aistudio.google.com
   • Sign in with your Google account
   • Navigate to API Keys section

2. Generate Veo 3 API Key

   • Click Create API Key
   • Select Google Veo 3 from available APIs
   • Copy the generated key
   • Note: Veo 3 may require waitlist access—check availability in your region

3. Set Up Billing

   • Veo 3 video generation incurs costs per video
   • Configure billing in Google Cloud Console
   • Set monthly budget limits to control expenses
   • Typical cost: $0.10-0.50 per video depending on length/quality

4. Configure in TaskAGI

   • Go to Integrations → Google Veo 3
   • Select Add New Integration
   • Paste your API key
   • Set video parameters:
     • Duration: 15-30 seconds (optimal for social media)
     • Resolution: 1080p (standard for TikTok/Instagram)
     • Frame rate: 24fps (cinematic quality)

5. Test Video Generation

   • Use a sample script to generate a test video
   • Verify output quality and duration
   • Check that videos download successfully

How to obtain API keys/credentials:

• Keys generated through Google AI Studio
• Veo 3 may have limited availability—join waitlist if needed
• Consider alternative video APIs if Veo 3 unavailable in your region

Configuration in TaskAGI:

• Store API key as {{secrets.veo3_api_key}}
• Set up cost monitoring to track video generation expenses
• Configure fallback options if API rate limits are exceeded

Configuration Steps

Node-by-Node Configuration Guidance

Node 1: Workflow Overview (core.workflow_note)

• Purpose: Documentation node explaining the workflow
• Configuration: Add a descriptive note:

  UGC Video Generation Workflow
  Converts product images → AI scripts → Professional videos
  Triggered via Telegram, delivered back to user
  

• No parameters to configure—informational only

Node 2: Receive Product Image (telegram.webhook)

• Purpose: Listens for incoming product images from Telegram users
• Configuration:
  • Trigger type: message_with_photo
  • Bot token: {{secrets.telegram_token}}
  • Expected input: Photo file from Telegram
• Data output: Image file ID, user ID, caption (if provided)
• Example: User sends product photo with caption "Blue wireless headphones"

Node 3: Extract Image & Caption (core.edit_data)

• Purpose: Parses incoming Telegram data and prepares it for processing
• Configuration:
  • Input data: Raw Telegram webhook payload
  • Extract fields:
    • photo_id → Store as {{photo_id}}
    • caption → Store as {{product_caption}}
    • user_id → Store as {{user_id}}
  • Transformation: Convert photo ID to downloadable format
• Example output:

  {
    "photo_id": "AgADAgADr6cxG...",
    "product_caption": "Blue wireless headphones",
    "user_id": "123456789"
  }
  

Node 4: Download Telegram Image (telegram.getFile)

• Purpose: Downloads the actual image file from Telegram servers
• Configuration:
  • File ID: {{photo_id}} (from Node 3)
  • Bot token: {{secrets.telegram_token}}
  • Output format: Binary image file
  • Storage: Temporary cache (auto-deleted after workflow completion)
• Parameter: file_path will be generated automatically

Node 5: Analyze Product & Generate Script (googlegemini.analyzeImage)

• Purpose: AI analyzes the product image and writes a compelling UGC ad script
• Configuration:
  • API key: {{secrets.gemini_api_key}}
  • Model: gemini-2.5-flash
  • Input image: {{downloaded_image}} (from Node 4)
  • System prompt:

    You are a professional UGC ad scriptwriter specializing in 
    high-converting product advertisements. Analyze the product 
    image and create a 30-second script that:
    - Opens with a relatable problem
    - Showcases product benefits naturally
    - Includes a clear call-to-action
    - Uses conversational, authentic language
    - Targets social media audiences (TikTok/Instagram style)
    

  • Temperature: 0.7 (balanced creativity)
  • Max tokens: 1500
• Example output:

  "You know that frustrating moment when your headphones 
  keep falling out during workouts? These wireless headphones 
  have this incredible secure-fit design... [continues for 30 seconds]"
  

Node 6: Generate UGC Video (googleveo3.generateVideo)

• Purpose: Creates a professional video from the AI-generated script
• Configuration:
  • API key: {{secrets.veo3_api_key}}
  • Input script: {{generated_script}} (from Node 5)
  • Input image: {{downloaded_image}} (from Node 4)
  • System prompt:

    Create a UGC-style product advertisement video for social media.
    - Duration: 15-30 seconds
    - Style: Authentic, relatable, high-energy
    - Include product shots, lifestyle imagery, and text overlays
    - Optimize for TikTok/Instagram Reels (9:16 aspect ratio)
    - Add trending background music and sound effects
    

  • Video parameters:
    • Duration: 30 seconds
    • Resolution: 1080x1920 (mobile-optimized)
    • Frame rate: 24 fps
    • Aspect ratio: 9:16 (vertical for social media)
• Processing time: 2-5 minutes per video
• Output: MP4 video file

Node 7: Send Video to User (telegram.sendVideo)

• Purpose: Delivers the finished video back to the user via Telegram
• Configuration:
  • Bot token: {{secrets.telegram_token}}
  • Chat ID: {{user_id}} (from Node 3)
  • Video file: {{generated_video}} (from Node 6)
  • Caption:

    ✨ Your UGC video is ready!
    
    Product: {{product_caption}}
    Generated with AI
    Ready to post on TikTok, Instagram, YouTube Shorts
    

  • Parse mode: HTML (for formatted text)
• Notification: User receives video instantly in Telegram

Nodes 8-9: Documentation Notes

• AI Script Generation and Video Generation: Informational nodes documenting each stage
• No configuration needed—reference only

Data Flow Between Nodes

The workflow follows this sequential data pipeline:

Telegram Image (Node 2)
    ↓
Extract metadata (Node 3)
    ↓
Download image file (Node 4)
    ↓
Analyze + Generate script (Node 5)
    ↓
Create video (Node 6)
    ↓
Send to user (Node 7)

Key data transformations:

• Node 2→3: Raw webhook → Structured data
• Node 3→4: Photo ID → Binary image file
• Node 4→5: Image file → Text script
• Node 5→6: Script + image → Video file
• Node 6→7: Video file → Telegram message

Testing Your Agent

Step 1: Pre-Launch Verification

Before running your first workflow:

1. Verify all integrations are connected

   • Go to Integrations dashboard
   • Confirm green checkmarks next to Telegram, Gemini, and Veo 3
   • Test each connection individually

2. Check API quotas and billing

   • Verify Gemini API has available quota
   • Confirm Veo 3 billing is active and funded
   • Set up cost alerts for Veo 3 usage

3. Validate webhook configuration

   • Send a test message to your Telegram bot
   • Verify it appears in TaskAGI activity logs
   • Confirm webhook URL is active

Step 2: Execute Test Workflow

1. Prepare a test product image

   • Use a clear, well-lit product photo
   • Include product in frame with context
   • Recommended: 1080x1080px or larger

2. Send test image to your bot

   • Open Telegram and find your bot
   • Send the product image with caption: "Test product: Blue wireless headphones"
   • Note the timestamp

3. Monitor workflow execution

   • In TaskAGI, go to Workflows → Executions
   • Find your test execution
   • Watch real-time progress through each node

Step 3: Verify Each Processing Stage

At Node 2 (Receive Image):

• ✅ Image received and logged
• ✅ User ID captured correctly
• ✅ Caption extracted (if provided)

At Node 3 (Extract Data):

• ✅ Photo ID extracted
• ✅ Metadata properly formatted
• ✅ No data loss or corruption

At Node 4 (Download Image):

• ✅ Image file successfully downloaded
• ✅ File size reasonable (typically 100KB-5MB)
• ✅ No timeout errors

At Node 5 (Generate Script):

• ✅ Gemini API responds within 30 seconds
• ✅ Script generated (typically 200-400 words)
• ✅ Script is coherent and product-focused
• ✅ No API errors or quota exceeded messages

At Node 6 (Generate Video):

• ✅ Veo 3 API accepts request
• ✅ Video generation starts (processing time: 2-5 minutes)
• ✅ Video file created (typically 5