AI Video Generator with Kling V2.1

What This Agent Does

This workflow automates the creation of professional AI-generated videos by combining intelligent prompt enhancement with cutting-edge video generation technology. When you submit a video request through a simple form, the agent intelligently processes your prompt—optionally enhancing it with GPT-4o's expertise—and then generates a high-quality video using Kling V2.1's advanced capabilities.

Key benefits include:

• Time savings: Reduce video creation from hours to minutes
• Quality improvement: AI-enhanced prompts produce more compelling, detailed video outputs
• Flexibility: Choose between using your original prompt or letting AI optimize it for better results
• Seamless workflow: Automated end-to-end process from request to video generation

Target use cases:

• Marketing teams creating product demonstration videos
• Content creators generating visual assets for social media
• Educational institutions producing instructional videos
• Creative professionals exploring AI-assisted video production
• Agencies scaling video content production without additional resources

Who Is It For

This workflow is ideal for content creators, marketing professionals, and creative teams who want to leverage AI for video production without deep technical knowledge. Whether you're a solo entrepreneur, a small marketing team, or part of a larger organization, this agent streamlines your video creation process while maintaining creative control through the optional prompt enhancement feature.

You'll benefit most if you:

• Regularly need to produce video content
• Want to experiment with AI-generated visuals
• Prefer automation over manual video editing
• Need quick turnaround times for video assets
• Want to maintain consistency in video quality

Required Integrations

OpenAI

Why it's needed: OpenAI's GPT-4o model powers the intelligent prompt enhancement feature. When enabled, it transforms your basic video descriptions into detailed, professionally-crafted prompts that generate superior video results. This AI-to-AI collaboration ensures your creative vision is fully realized.

Setup steps:

1. Create an OpenAI account at https://platform.openai.com
2. Navigate to API keys in your account settings (Profile → API keys)
3. Generate a new secret key by clicking "Create new secret key"
4. Copy the key immediately (you won't be able to view it again)
5. In TaskAGI, go to Integrations → OpenAI
6. Paste your API key in the authentication field
7. Enable the integration and verify the connection status shows "Connected"

How to obtain API keys:

• Visit your OpenAI account dashboard
• Ensure you have billing set up (GPT-4o requires a paid account)
• Create an API key with appropriate permissions
• Store the key securely—never share it publicly

Configuration in TaskAGI:

• Model: Ensure it's set to gpt-4o (the most capable model for prompt engineering)
• Temperature: Keep at 0.7 for balanced creativity and consistency
• Max tokens: Set to 1000 to allow detailed prompt generation
• Test connection: Use the "Test Integration" button to verify setup

Kling V2.1

Why it's needed: Kling V2.1 is the core video generation engine that transforms your prompts into stunning AI-generated videos. This state-of-the-art model produces high-quality, coherent video content with excellent visual fidelity and motion quality.

Setup steps:

1. Sign up for Kling AI at https://klingai.com
2. Complete account verification and email confirmation
3. Access your API dashboard from the account settings
4. Generate API credentials (API key and secret)
5. Copy both the API key and secret to a secure location
6. In TaskAGI, navigate to Integrations → Kling V2.1
7. Enter your API key in the designated field
8. Enter your API secret in the corresponding field
9. Save and test the connection

How to obtain API keys/credentials:

• Log into your Kling AI account
• Go to Developer Settings or API Management
• Create a new application/project
• Generate credentials with video generation permissions
• Note any rate limits or quota information for your account tier

Configuration in TaskAGI:

• API Key: Your primary authentication credential
• API Secret: Your secondary authentication credential
• Video Duration: Set to 60 seconds (adjustable based on your needs)
• Resolution: Select 1080p for optimal quality
• Frame Rate: Keep at 24fps for smooth motion
• Test generation: Run a test video to confirm the integration works

Configuration Steps

Step 1: Overview Node (core.workflow_note)

This is your workflow's documentation hub. While not functional, it's crucial for team collaboration.

What to do:

• Add a clear title: "AI Video Generation Workflow"
• Write a brief description explaining the workflow's purpose
• Include any important notes about enhancement mode or expected processing time
• Document any team guidelines for prompt submission

Example note:

This workflow generates professional AI videos from text prompts.
Enable Enhancement Mode for optimized results.
Processing typically takes 2-5 minutes per video.

Step 2: Video Request Form (trigger.form)

This is where users submit their video requests. Configure it to capture all necessary information.

Form fields to create:

1. Video Prompt (Text Area, Required)

   • Label: "Describe your video concept"
   • Placeholder: "A serene mountain landscape with flowing clouds at sunset"
   • Help text: "Be descriptive but concise. Include style, mood, and key elements."

2. Enhancement Mode (Toggle/Checkbox, Required)

   • Label: "Enhance prompt with AI?"
   • Default: true (enabled)
   • Help text: "AI will optimize your prompt for better video quality"

3. Video Style (Dropdown, Optional)

   • Options: Cinematic, Realistic, Artistic, Abstract, Documentary
   • Help text: "Helps guide the video generation style"

4. Submitter Email (Email, Optional)

   • Label: "Your email for notifications"
   • Help text: "We'll send you the video link when ready"

Step 3: Check Enhancement Mode (core.switch)

This decision node routes the workflow based on whether the user enabled AI enhancement.

Configuration:

• Switch condition: Check if enhancement_mode equals true
• True path: Routes to GPT-4o enhancement node
• False path: Routes directly to "Set Original Prompt" node
• Both paths merge at the "Merge Prompt Paths" node

Why this matters: Users who want quick results can skip enhancement; those seeking optimized prompts benefit from AI refinement.

Step 4: Enhance Prompt with GPT-4o (openai.createCompletion)

This node transforms basic prompts into detailed, professional video descriptions.

Configuration:

Model: gpt-4o
Temperature: 0.7
Max Tokens: 1000

System Prompt (already configured):

You are an expert AI video prompt engineer. Transform the user's 
basic video concept into a detailed, vivid prompt optimized for 
AI video generation. Include specific visual details, camera movements, 
lighting, mood, and cinematographic elements. Keep it under 500 words.

User Input: Map this to the form's video_prompt field

Expected output: A detailed, enhanced prompt ready for video generation

Step 5: Set Enhanced Prompt (core.edit_data)

This node prepares the enhanced prompt for the merge operation.

Configuration:

• Input data: Connect to GPT-4o completion output
• Extract field: choices[0].message.content (the actual prompt text)
• Output variable name: final_prompt
• Add metadata: Include source: "enhanced"

Step 6: Set Original Prompt (core.edit_data)

This node handles the non-enhancement path, preparing the original prompt.

Configuration:

• Input data: Connect to form submission data
• Extract field: video_prompt (user's original text)
• Output variable name: final_prompt
• Add metadata: Include source: "original"

Step 7: Merge Prompt Paths (core.merge)

This node combines both paths, ensuring consistent data structure regardless of enhancement choice.

Configuration:

• Merge strategy: Combine all fields
• Conflict resolution: Keep the final_prompt field from whichever path executed
• Output structure: Ensure final_prompt is accessible to the next node

Step 8: Generate Video with Kling V2.1 (klingv21.generateVideo)

This is the core video generation node where your prompt becomes reality.

Configuration:

Prompt: [[nodes.7629.merged_data.final_prompt]]
Duration: 60 seconds
Resolution: 1080p
Frame Rate: 24fps
Quality: High

Parameters explained:

• Prompt: Uses the merged final_prompt from the previous node
• Duration: Adjust based on your needs (30-120 seconds typical)
• Resolution: 1080p balances quality and processing time
• Frame Rate: 24fps is cinema standard; 30fps for smoother motion

Expected output: Video file URL and metadata

Step 9: Prepare Response (core.edit_data)

This final node formats the output for user delivery.

Configuration:

• Extract video URL from Kling output
• Include metadata: Generation timestamp, prompt used, enhancement status
• Format response as JSON or user-friendly message
• Add download link and preview information

Example output structure:

{
  "video_url": "https://...",
  "video_title": "Generated Video",
  "prompt_used": "...",
  "enhanced": true,
  "generation_time": "3m 45s",
  "status": "completed"
}

Testing Your Agent

Test Execution Steps

1. Initial Setup Test

• Click "Test Workflow" from the workflow editor
• Verify all integrations show "Connected" status
• Confirm no configuration errors appear

2. Form Submission Test

• Submit a test form with a simple prompt: "A golden retriever running through a field of sunflowers"
• Enable enhancement mode
• Note the submission timestamp

3. Enhancement Verification

• Monitor the GPT-4o node execution
• Verify the enhanced prompt appears more detailed than the original
• Check that the prompt maintains your original concept while adding visual depth

4. Video Generation Test

• Confirm the Kling V2.1 node receives the final prompt
• Monitor generation progress (typically 2-5 minutes)
• Verify the video URL appears in the response

5. Output Validation

• Download the generated video
• Check video quality and duration
• Verify it matches your prompt's intent
• Review metadata in the response

Success Indicators

✅ All nodes execute without errors ✅ Enhancement mode produces noticeably detailed prompts ✅ Video generates within expected timeframe ✅ Video quality matches your standards ✅ Response includes all necessary metadata ✅ Form submissions are logged correctly

Troubleshooting Tips

• Slow generation: Check your Kling API rate limits
• Poor video quality: Refine your prompt with more specific details
• Enhancement not working: Verify OpenAI API key has sufficient credits
• Form not submitting: Check required field validation settings

Your AI video generation workflow is now ready to transform creative ideas into stunning visual content!