Skip to content

AI_WORKFLOW_GUIDE.md

Latest commit

 

History

History
344 lines (257 loc) · 8.18 KB

AI_WORKFLOW_GUIDE.md

File metadata and controls

344 lines (257 loc) · 8.18 KB

AI Workflow Building Guide

This guide shows you how to use the AI Assistant to build generative media workflows in Clinks through natural language commands.

🎯 Overview

The AI Assistant can help you:

  • Add individual nodes to your canvas
  • Build complete multi-node workflows with automatic connections
  • Connect nodes together to create custom data flows
  • Execute workflows to generate content
  • Check workflow status and monitor progress

🚀 Getting Started

1. Open the AI Assistant

Click the "AI ASSISTANT" button at the bottom-center of the screen to open the chat interface.

2. Start Building

Simply tell the AI what you want to create. The AI understands natural language and will:

  • Choose the right node types
  • Position them intelligently
  • Connect them in the correct order
  • Provide visual feedback as it builds

💬 Example Commands

Adding Single Nodes

Add individual nodes one at a time:

"Add a text input node"
"Create an image generator"
"Add a video generator node"
"I need a logo generator"
"Add an output node"

Node Types Available:

  • text-input - For entering text prompts
  • image-input - For uploading reference images
  • text-generator - For AI text generation
  • image-generator - For creating AI images
  • video-generator - For creating AI videos
  • logo-generator - For generating brand logos
  • audio-generator - For creating AI audio
  • output - For displaying and exporting results

Building Complete Workflows

Create entire pipelines with one command:

"Build a text-to-image workflow"
"Create a logo generation pipeline"
"Set up a video creation workflow with text input and output"
"Build a workflow that takes text and generates an image"

The AI will:

  1. Determine the required nodes
  2. Add them in the correct order
  3. Automatically connect them
  4. Position them nicely on the canvas

Checking Workflow Status

See what's in your current workflow:

"Show my workflow status"
"What nodes do I have?"
"What's in my workflow?"
"How many nodes are there?"

Executing Workflows

Run your workflow to generate content:

"Execute the workflow"
"Run my workflow"
"Generate content"
"Execute this pipeline"

🏗️ Common Workflow Patterns

Text-to-Image Workflow

"Build a text-to-image workflow"

This creates:

  1. Text Input - Where you enter your prompt
  2. Image Generator - Generates the image from text
  3. Output - Displays the generated image

Logo Generation Workflow

"Create a logo generation workflow"

This creates:

  1. Text Input - Describe the logo you want
  2. Logo Generator - Creates the logo
  3. Output - Shows the generated logo

Video Creation Workflow

"Build a video generation pipeline"

This creates:

  1. Text Input - Video description/prompt
  2. Video Generator - Creates the video
  3. Output - Displays the video

Multi-Stage Workflow

"Build a workflow that generates text, then uses that text to create an image"

This creates:

  1. Text Input - Initial prompt
  2. Text Generator - Generates descriptive text
  3. Image Generator - Creates image from generated text
  4. Output - Shows the final image

🔧 Advanced Usage

Custom Positioning

You can specify where nodes should appear:

"Add an image generator at position 500, 300"

Multiple Nodes at Once

Ask for several nodes in one request:

"Add a text input, an image generator, and an output node"

Custom Connections

Build nodes first, then connect them specifically:

1. "Add a text input called 'Prompt A'"
2. "Add a text input called 'Prompt B'"
3. "Add an image generator"
4. "Connect both text inputs to the image generator"

Workflow Templates

Ask for common patterns:

"Create a basic generation workflow"
"Set up a standard image pipeline"
"Build a complete video workflow with all the nodes I need"

📊 Understanding Workflow Execution

When you execute a workflow:

  1. Input Nodes are processed first (text-input, image-input)
  2. Generator Nodes receive data from inputs and create content
  3. Output Nodes display the final results
  4. Data flows along the connections you've created

Before Execution

Make sure:

  • ✅ All input nodes have content
  • ✅ Generator nodes are properly configured
  • ✅ Nodes are connected in the right order
  • ✅ There's an output node to see results

During Execution

  • Watch the Preview Panel for real-time results
  • Execution logs show progress for each node
  • Errors will be displayed if something goes wrong

After Execution

  • Check the Preview Panel for generated content
  • Click on output nodes to see final results
  • Review execution logs for any issues

🎨 Tips for Best Results

1. Start Simple

Begin with basic workflows and add complexity:

"Build a simple text-to-image workflow"

2. Be Specific

Clear instructions help the AI choose the right nodes:

"Create a workflow for generating product logos with text input and output"

3. Build Incrementally

Add and test nodes step by step:

1. "Add a text input"
2. "Add an image generator"
3. "Show my workflow status"
4. "Connect them together"

4. Use Natural Language

Don't worry about exact syntax:

"I want to make images from text descriptions"
"Help me build something that generates videos"
"Set me up to create logos"

5. Ask for Help

The AI can guide you:

"What kind of workflow should I build for creating marketing images?"
"How do I set up a video generation pipeline?"
"What nodes do I need for text-to-image generation?"

🔍 Troubleshooting

"No nodes in workflow"

Solution: Add nodes first before trying to execute

"Build a text-to-image workflow"

"Nodes not connected"

Solution: Make sure nodes are connected

"Connect my nodes together"

"No input provided"

Solution: Configure input nodes with content before executing

"Component not rendering"

Problem: AI response doesn't show a component Solution: Try rephrasing your request more specifically:

  • Instead of: "add node"
  • Try: "add a text generator node"

📚 Workflow Examples Library

Example 1: Basic Image Generation

User: "Build a text-to-image workflow"
AI: Creates text-input → image-generator → output
User: "Execute the workflow"

Example 2: Logo Design

User: "Create a logo generation pipeline"
AI: Creates text-input → logo-generator → output

Example 3: Video Creation

User: "Set up a video creation workflow"
AI: Creates text-input → video-generator → output

Example 4: Audio Generation

User: "Build an audio generation workflow"
AI: Creates text-input → audio-generator → output

Example 5: Complex Multi-Stage

User: "Build a workflow that generates a description, then creates an image, then shows the output"
AI: Creates text-input → text-generator → image-generator → output

🎯 Best Practices

✅ DO:

  • Use clear, descriptive node names
  • Build workflows step by step when learning
  • Check workflow status before executing
  • Test with simple inputs first
  • Ask the AI for help when stuck

❌ DON'T:

  • Try to execute empty workflows
  • Skip connecting nodes in custom workflows
  • Forget to configure generator nodes
  • Ignore error messages
  • Be afraid to experiment!

🆘 Getting Help

If you're stuck, ask the AI:

"How do I create a text-to-image workflow?"
"What's wrong with my workflow?"
"Help me set up image generation"
"Show me an example workflow"
"What can I build with this?"

🚀 Next Steps

  1. Try the examples - Start with simple workflows
  2. Experiment - Mix and match different node types
  3. Share - Export your workflows for others
  4. Build - Create custom pipelines for your needs

📖 Related Documentation

Remember: The AI Assistant is here to help! Don't hesitate to ask questions, try different approaches, or request clarification. Building workflows should be intuitive and fun! 🎉