n8n-nodes-comfyui-all 2.4.5 → 2.4.7

package/README.md

@@ -1,319 +1,382 @@
 # n8n-nodes-comfyui-all
 
-> n8n community node for integrating ComfyUI workflows into n8n automation platform.
+> Execute ComfyUI workflows in n8n - Generate images, videos, and more with AI!
 
 [![npm version](https://badge.fury.io/js/n8n-nodes-comfyui-all.svg)](https://www.npmjs.com/package/n8n-nodes-comfyui-all)
 
-## ✨ Features
+## Video Tutorials
 
-- 🎨 **Universal Workflow Support** - Works with any ComfyUI workflow in API format
-- 🔄 **Dynamic Parameters** - Override workflow parameters dynamically
-- 🎬 **Multi-Modal Support** - Supports images and videos for both input and output
-- 🤖 **AI Agent Ready** - Two specialized nodes for different use cases:
-  - **ComfyUI Tool** - Optimized for AI Agent workflows (URL-based image input)
-  - **ComfyUI** - Full-featured node for standard workflows (binary & URL support)
-- 📊 **Flexible Configuration** - JSON mode or single parameter mode
-- 🔗 **Multiple Image Input Methods** - URL, binary data
-- 🏷️ **Customizable Output** - Customize binary output property names
+- 📺 [YouTube Tutorial](https://youtu.be/wsbo3hBKsPM)
+- 📺 [哔哩哔哩教程](https://www.bilibili.com/video/BV1ffrFBTEdQ/?vd_source=6485fe2fae664d8b09cb2e2fd7df5ef7)
 
-## 📦 Installation
+## What This Does
 
-### n8n Cloud
+This package adds **two nodes** to n8n that let you run ComfyUI workflows:
 
-1. Go to **Settings** → **Community Nodes**
-2. Click **Install**
-3. Enter: `n8n-nodes-comfyui-all`
-4. Click **Install**
+1. **ComfyUI Tool** - For AI Agents (URL-based image input)
+2. **ComfyUI** - For standard workflows (supports binary & URL)
 
-### Self-Hosted n8n
+You can use these nodes to:
+- Generate images from text prompts
+- Process and edit images
+- Generate videos
+- Use AI Agents to create images automatically
+- And much more with any ComfyUI workflow!
 
-```bash
-cd ~/.n8n
-npm install n8n-nodes-comfyui-all
-```
-
-Or via n8n interface: **Settings** → **Community Nodes** → **Install** → `n8n-nodes-comfyui-all`
+---
 
-> Restart n8n after installation.
+## Quick Setup
 
-## 🚀 Quick Start
+### 1. Install
 
-### 1. Prerequisites
+```bash
+# n8n Cloud: Settings → Community Nodes → Install → n8n-nodes-comfyui-all
 
-- ComfyUI server running (default: `http://127.0.0.1:8188`)
-- n8n instance (version 2.x or higher)
+# Self-hosted:
+cd ~/.n8n
+npm install n8n-nodes-comfyui-all
+```
 
-### 2. Create Your Workflow in ComfyUI
+### 2. Run ComfyUI
 
-1. Design your workflow in ComfyUI
-2. Click **Save (API Format)** to export
-3. Copy the generated JSON
+Make sure ComfyUI is running:
+```bash
+# Default: http://127.0.0.1:8188
+```
 
-### 3. Configure n8n Node
+### 3. Use in n8n
 
-1. Add **ComfyUI** node to your n8n workflow
-2. Set **ComfyUI URL**: `http://127.0.0.1:8188`
-3. Paste your **Workflow JSON**
-4. Optionally configure **Node Parameters**
+Add either **ComfyUI Tool** or **ComfyUI** node to your workflow!
 
-## 📖 Usage
+---
 
-### Basic Example: Text to Image
+## ComfyUI Tool vs ComfyUI Node
 
-**Node Configuration:**
-- **ComfyUI URL**: `http://127.0.0.1:8188`
-- **Workflow JSON**: Your ComfyUI workflow in API format
+### ComfyUI Tool 🤖
 
-**Optional Node Parameters (Single Parameter Mode):**
-- Node ID: `6` (your CLIP text node)
-- Parameter Name: `text`
-- Type: `Text`
-- Value: `a beautiful landscape, high quality`
+**Use for: AI Agent workflows**
 
-### Parameter Configuration Modes
+```
+Example:
+[Chat Input] → [AI Agent] → [ComfyUI Tool]
+```
 
-#### 1. Multiple Parameters Mode (JSON)
+**Features:**
+- ✅ Simple URL-based image input
+- ✅ Works great with AI Agents
+- ✅ Doesn't bloat LLM context (URLs are small!)
+- ✅ Automatic image download
 
-Configure multiple parameters at once using JSON:
+**How to use:**
+1. Add to AI Agent as a tool
+2. Set ComfyUI URL
+3. Paste your ComfyUI workflow JSON
+4. Set Load Image Node ID (if using images)
+5. Pass image URL: `{{ $json.imageUrl }}`
 
+**Example:**
 ```json
 {
-  "width": 1024,
-  "height": 1024,
-  "batch_size": 1,
-  "seed": 12345
+  "imageUrl": "https://example.com/image.png",
+  "prompt": "Transform to oil painting"
 }
 ```
 
-#### 2. Single Parameter Mode
-
-Configure one parameter at a time with type validation:
+---
 
-- **Parameter Name**: `steps`
-- **Type**: `Number`
-- **Value**: `25`
+### ComfyUI Node 🔧
 
-### Image Input Options
+**Use for: Standard n8n workflows**
 
-The node supports two methods for uploading images to ComfyUI:
+```
+Example:
+[HTTP Request] → [ComfyUI] → [Save File]
+```
 
-#### Method 1: Binary Data
+**Features:**
+- ✅ Supports binary data (from previous nodes)
+- ✅ Supports URL download
+- ✅ Advanced parameter configuration
+- ✅ Binary output for further processing
 
-Upload images from n8n binary data:
+**How to use:**
 
-- **Type**: `Image`
-- **Image Input Type**: `Binary`
-- **Value**: Binary property name (e.g., `data`, `image`, `file`)
+**Option 1: Binary Input**
+```
+ComfyUI URL: http://127.0.0.1:8188
+Workflow JSON: [Your workflow]
+
+Node Parameters:
+  Node ID: 107
+  Type: Image
+  Image Input Type: Binary
+  Value: data
+```
 
-#### Method 2: URL
+**Option 2: URL Input**
+```
+ComfyUI URL: http://127.0.0.1:8188
+Workflow JSON: [Your workflow]
+
+Node Parameters:
+  Node ID: 107
+  Type: Image
+  Image Input Type: URL
+  Image URL: https://example.com/image.png
+```
 
-Download and upload images from a URL:
+**Option 3: Text Input**
+```
+Node Parameters:
+  Node ID: 6
+  Type: Text
+  Value: {{ $json.prompt }}
+```
 
-- **Type**: `Image`
-- **Image Input Type**: `URL`
-- **Image URL**: Full URL of the image (e.g., `https://example.com/image.png`)
+---
 
-The node will automatically download the image from the URL and upload it to ComfyUI.
+## Common Workflows
 
-### Custom Output Binary Key
+### 1. Generate Image from Text
 
-By default, the first output image/video uses `data` as the binary property name. You can customize this:
+**Setup:**
+```
+ComfyUI Node:
+  ComfyUI URL: http://127.0.0.1:8188
+  Workflow JSON: [Text-to-Image workflow]
+
+  Node Parameters:
+    Node ID: 6 (your CLIP Text node)
+    Type: Text
+    Value: {{ $json.prompt }}
+```
 
-- **Output Binary Key**: Property name for the first output (e.g., `image`, `output`, `result`)
-- Default: `data`
+**Input:**
+```json
+{ "prompt": "a beautiful sunset over mountains" }
+```
 
-**Example:**
+**Output:**
 ```json
 {
-  "binary": {
-    "myImage": { "data": "...", "mimeType": "image/png" }
-  }
+  "imageUrls": ["http://127.0.0.1:8188/view?filename=image.png"]
 }
 ```
 
-## 🤖 AI Agent Integration
-
-Use ComfyUI as a tool in AI Agent workflows.
-
-### Quick Setup
-
-**Step 1: Create AI Agent**
-1. Add **OpenAI Conversational Agent** node
-2. Configure Chat Model (GPT-4/GPT-3.5)
-3. Add Memory (optional but recommended)
-
-**Step 2: Add ComfyUI Tool**
-1. Click AI Agent node
-2. In **Tools** section, click **+ Add Tool**
-3. Search and select **ComfyUI**
-4. Configure:
-   - **ComfyUI URL**: `http://127.0.0.1:8188`
-   - **Workflow JSON**: Your ComfyUI workflow (API format)
-   - **Node Parameters**: Configure text parameter override
+---
 
-**Step 3: Start Chatting**
-Execute workflow and start conversing!
+### 2. Process Image with AI Agent
 
-### Example Conversations
-
-**Basic Image Generation:**
-
-**User:**
+**Workflow:**
 ```
-Generate a picture of a cute cat sitting on a fence
+[Chat Interface] → [AI Agent] → [ComfyUI Tool]
 ```
 
-**AI:**
+**ComfyUI Tool Setup:**
 ```
-I'll generate that for you.
-[Executes ComfyUI with prompt: "a cute cat sitting on a fence"]
-Done! Here's the image.
+ComfyUI URL: http://127.0.0.1:8188
+Workflow JSON: [Image-to-Image workflow]
+Load Image Node ID: 107
+Image URL: {{ $json.imageUrl }}
 ```
 
-**With Parameters:**
-
-**User:**
+**Chat:**
 ```
-Create a cyberpunk city, size:1024x768, steps:30
+You: Transform this image https://example.com/photo.png
+     to watercolor style
+
+AI: I'll transform that image for you!
+    [Downloads image, processes in ComfyUI]
+    Done! Here's your watercolor image 🎨
 ```
 
-**Supported Parameters:**
-- `size:WIDTHxHEIGHT` - Image dimensions
-- `steps:N` - Sampling steps
-- `cfg:N` - CFG strength
-- `seed:N` - Random seed
-- `negative:TEXT` - Negative prompt
+---
 
-**Negative Prompts:**
+### 3. Edit Image with Parameters
 
-**User:**
+**ComfyUI Node Setup:**
 ```
-Draw a beautiful sunset, negative: blurry, low quality
+Node Parameters:
+  Node ID: 3
+  Parameter Mode: Multiple Parameters
+  Parameters JSON:
+    {
+      "width": 1024,
+      "height": 1024,
+      "steps": 30
+    }
 ```
 
-### Configuring Node Parameters for AI Agent
+---
 
-In the ComfyUI node, configure parameter overrides:
+### 4. Generate Video
+
+Same as image generation, just use a video workflow in ComfyUI!
+
+**Output:**
+```json
+{
+  "videoUrls": ["http://127.0.0.1:8188/view?filename=video.mp4"],
+  "videoCount": 1
+}
+```
 
-- **Node ID**: `6` (your CLIP text node)
-- **Parameter Mode**: Single Parameter
-- **Parameter Name**: `text`
-- **Type**: Text
-- **Value**: [Leave empty - AI Agent will fill this]
+---
 
-### Tool Mode Output Format
+## How to Get Your Workflow JSON
 
-When used as an AI Agent tool, ComfyUI returns both binary data and URL information:
+1. Open ComfyUI
+2. Create or load your workflow
+3. Click **"Save (API Format)"** (not "Save")
+4. Copy the JSON
+5. Paste it in the node's **Workflow JSON** field
 
-**Binary Output:**
-- Images are returned as binary data (base64-encoded) in the `binary` object
-- First image: `binary.data`
-- Additional images: `binary.image_1`, `binary.image_2`, etc.
-- Videos: `binary.video_0`, `binary.video_1`, etc.
+---
 
-**JSON Output (with URLs):**
-- `json.images`: Array of image paths
-- `json.imageUrls`: Array of complete image URLs (e.g., `http://127.0.0.1:8188/view?filename=...`)
-- `json.videos`: Array of video paths
-- `json.videoUrls`: Array of complete video URLs
-- `json.imageCount`: Number of images generated
-- `json.videoCount`: Number of videos generated
+## Parameters Explained
 
-This ensures compatibility with both workflow mode (binary data) and tool mode (URLs for AI Agent).
+### ComfyUI Tool Node
 
-### Best Practices
+| Parameter | What It Does | Example |
+|-----------|--------------|---------|
+| **ComfyUI URL** | Where ComfyUI is running | `http://127.0.0.1:8188` |
+| **Workflow JSON** | Your ComfyUI workflow (API format) | `{...}` |
+| **Timeout** | Max wait time (seconds) | `300` |
+| **Load Image Node ID** | Which LoadImage node to use | `107` |
+| **Image URL** | URL of image to process | `https://...` |
+| **Parameter Overrides** | Change workflow values | See below |
 
-**1. Clear Prompts**
+### ComfyUI Node
 
-**Good:** "Generate a landscape painting of mountains at sunset"
+| Parameter | What It Does | Example |
+|-----------|--------------|---------|
+| **ComfyUI URL** | Where ComfyUI is running | `http://127.0.0.1:8188` |
+| **Workflow JSON** | Your ComfyUI workflow (API format) | `{...}` |
+| **Timeout** | Max wait time (seconds) | `300` |
+| **Output Binary Key** | Name for output binary | `data` |
+| **Node Parameters** | Configure workflow parameters | See below |
 
-**Bad:** "Make a picture"
+### Node Parameters
 
-**2. Use Specific Parameters**
+| Field | What It Does |
+|-------|--------------|
+| **Node ID** | The ComfyUI node to change (e.g., "6", "13") |
+| **Parameter Mode** | "Single" for one parameter, "Multiple" for JSON |
+| **Type** | Text, Number, Boolean, or Image |
+| **Value** | The value to set |
 
-**Good:** "Create a portrait, size:512x768, steps:25"
+---
 
-**Bad:** "Create a high-quality portrait"
+## Tips & Tricks
 
-**3. Negative Prompts**
+### ✅ Use URLs with AI Agents
 
-Always use negative prompts for better quality:
+URLs are much smaller than base64 images:
+```
+Base64:  ~100KB - 1MB per image
+URL:     ~100 bytes
 
+Better for LLM context and costs!
 ```
-A beautiful landscape, negative: blurry, distorted, low quality
+
+### ✅ Find Your Node ID
+
+Open your workflow JSON and look for:
+```json
+{
+  "6": {
+    "inputs": {...},
+    "class_type": "KSampler"
+  }
+}
 ```
+The `"6"` is your Node ID!
+
+### ✅ Test First
 
-### Advanced Usage
+Always test your workflow in ComfyUI before using it in n8n!
 
-**Style Transfer:**
+### ✅ Use Appropriate Timeouts
 
-**User:**
 ```
-Transform this image to oil painting style
+Simple workflows:  60-120 seconds
+Complex workflows: 300-600 seconds
+Video generation:  600-1800 seconds
 ```
 
-Setup:
-- Upload input image as binary data
-- Configure style parameters
+---
+
+## Troubleshooting
+
+### "Invalid ComfyUI URL"
+Make sure ComfyUI is running and URL is correct: `http://127.0.0.1:8188`
+
+### "Workflow execution timeout"
+Increase the timeout value in node settings
+
+### "Node ID not found"
+Check your workflow JSON - Node IDs are strings like "6", not numbers
+
+### "Failed to download image"
+Make sure the URL is public and accessible (not localhost or file://)
+
+### "AI Agent doesn't call ComfyUI"
+- Make sure ComfyUI Tool is added to Agent's tools
+- Check the workflow JSON is set
+- Try being more specific in your chat
 
-**Multi-Modal Workflows:**
+---
 
-Combine ComfyUI with other tools:
+## Examples
 
-**User:**
+### Example 1: Simple Text to Image
+
+```
+[Manual Trigger] → [ComfyUI] → [Save to File]
+
+ComfyUI Node:
+  Workflow: Text-to-Image
+  Node ID: 6
+  Type: Text
+  Value: a cute cat in a garden
 ```
-Generate an image of a futuristic city, then write a poem about it
+
+### Example 2: AI Image Generator
+
 ```
+[Chat Trigger] → [AI Agent + ComfyUI Tool] → [Display Image]
 
-AI Agent:
-1. Calls ComfyUI to generate image
-2. Calls Chat Model to write poem
-3. Returns both results
+Agent says: "Generate a picture of..."
+Calls: ComfyUI Tool
+Returns: Image URL
+Shows: Result to user
+```
 
-### Troubleshooting
+### Example 3: Batch Process Images
 
-**AI Agent Doesn't Call ComfyUI**
+```
+[Spreadsheet] → [Loop] → [ComfyUI] → [Save]
 
-**Check:**
-1. ComfyUI tool added to Tools list?
-2. Workflow JSON configured?
-3. ComfyUI server running?
+For each row:
+  - Get image URL
+  - Process in ComfyUI
+  - Save result
+```
 
-**Images Don't Match Prompts**
+---
 
-**Check:**
-1. Correct node ID (e.g., `6` for CLIP text)
-2. Parameter name matches workflow (`text`)
-3. Workflow uses dynamic text input
+## Need Help?
 
-## 🔧 Configuration Reference
+- 📖 [n8n Documentation](https://docs.n8n.io)
+- 💬 [n8n Community](https://community.n8n.io)
+- 🐛 [Report Issues](https://github.com/your-repo/issues)
 
-| Field | Description |
-|-------|-------------|
-| **ComfyUI URL** | URL of your ComfyUI server |
-| **Workflow JSON** | ComfyUI workflow in API format |
-| **Timeout** | Maximum wait time in seconds (default: 300) |
-| **Output Binary Key** | Property name for first output binary data (default: `data`) |
-| **Node Parameters** | Override workflow parameters |
+---
 
-#### Node Parameter Fields
+## License
 
-| Field | Description |
-|-------|-------------|
-| **Node ID** | The node ID in your workflow (e.g., `6`, `13`) |
-| **Parameter Mode** | Single Parameter or Multiple Parameters (JSON) |
-| **Type** | Data type: Text, Number, Boolean, or Image |
-| **Image Input Type** | (When Type=Image) Binary or URL |
-| **Image URL** | (When Image Input Type=URL) URL to download image from |
-| **Parameter Name** | (Single mode) Parameter name to override |
-| **Parameters JSON** | (Multiple mode) JSON object with parameters |
-| **Value** | Value to set (varies by Type) |
+MIT
 
-## 💡 Tips
+---
 
-- **First time**: Start with a simple text-to-image workflow
-- **Parameter overrides**: Use Node Parameters instead of modifying workflow JSON
-- **Seed control**: Fixed seeds produce reproducible results
-- **Optimization**: 20-30 sampling steps are usually sufficient
-- **Image input**: Use Image type with Binary source for n8n binary data, or URL source to download from web
-- **Output naming**: Customize Output Binary Key to match your workflow's expected property names
+**Happy automating with ComfyUI! 🚀**

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "n8n-nodes-comfyui-all",
-  "version": "2.4.5",
+  "version": "2.4.7",
   "description": "n8n community nodes for ComfyUI workflow execution with dynamic parameter support",
   "keywords": [
     "n8n-community-node-package",