Resize, sharpen, and compress an image to fit email platform size limits in a single pipeline.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Generate a branded property listing graphic with a property photo, status badge, price, address, and key stats.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
설치하기 전에 에이전트에게 전달되는 지시 내용을 확인할 수 있습니다.
Real estate agencies and property platforms use this recipe to generate listing graphics for social media, email campaigns, and digital signage. Feed in property data and a photo URL to get a print-ready branded card — no design software, no manual exports.
Image Generation (1 credits/request)
You need an Iteration Layer API key. Get one at platform.iterationlayer.com during the 7-day trial.
For full integration guidance (SDKs, auth, MCP, error handling), see the Iteration Layer Integration Guide.
curl -X POST \
https://api.iterationlayer.com/image-generation/v1/generate \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"dimensions": { "width_in_px": 1080, "height_in_px": 1080 },
"output_format": "jpeg",
"layers": [
{
"index": 0,
"type": "image",
"file": {
"type": "url",
"name": "property.jpg",
"url": "https://example.com/properties/412-ridgewood.jpg"
},
"dimensions": { "width_in_px": 1080, "height_in_px": 720 }
},
{
"index": 1,
"type": "solid-color",
"hex_color": "#1A1A2E",
"position": { "x_in_px": 0, "y_in_px": 720 },
"dimensions": { "width_in_px": 1080, "height_in_px": 360 }
},
{
"index": 2,
"type": "solid-color",
"hex_color": "#E63946",
"position": { "x_in_px": 40, "y_in_px": 40 },
"dimensions": { "width_in_px": 180, "height_in_px": 48 },
"border_radius": 6
},
{
"index": 3,
"type": "text",
"text": "JUST LISTED",
"font_name": "Inter",
"font_size_in_px": 18,
"font_weight": "bold",
"text_color": "#FFFFFF",
"text_align": "center",
"vertical_align": "center",
"position": { "x_in_px": 40, "y_in_px": 40 },
"dimensions": { "width_in_px": 180, "height_in_px": 48 }
},
{
"index": 4,
"type": "text",
"text": "EUR 685,000",
"font_name": "Inter",
"font_size_in_px": 52,
"font_weight": "bold",
"text_color": "#FFFFFF",
"text_align": "left",
"vertical_align": "center",
"position": { "x_in_px": 40, "y_in_px": 740 },
"dimensions": { "width_in_px": 700, "height_in_px": 72 }
},
{
"index": 5,
"type": "text",
"text": "412 Ridgewood Lane, Berlin 10115",
"font_name": "Inter",
"font_size_in_px": 22,
"font_weight": "regular",
"text_color": "#CBD5E1",
"text_align": "left",
"vertical_align": "center",
"position": { "x_in_px": 40, "y_in_px": 820 },
"dimensions": { "width_in_px": 900, "height_in_px": 36 }
},
{
"index": 6,
"type": "text",
"text": "4 bed · 2.5 bath · 2,140 sq ft",
"font_name": "Inter",
"font_size_in_px": 20,
"font_weight": "regular",
"text_color": "#94A3B8",
"text_align": "left",
"vertical_align": "center",
"position": { "x_in_px": 40, "y_in_px": 870 },
"dimensions": { "width_in_px": 900, "height_in_px": 36 }
},
{
"index": 7,
"type": "text",
"text": "Hanse Realty GmbH",
"font_name": "Inter",
"font_size_in_px": 18,
"font_weight": "semibold",
"text_color": "#64748B",
"text_align": "right",
"vertical_align": "center",
"position": { "x_in_px": 40, "y_in_px": 1020 },
"dimensions": { "width_in_px": 1000, "height_in_px": 36 }
}
]
}'
import { IterationLayer } from "iterationlayer";
const client = new IterationLayer({ apiKey: "YOUR_API_KEY" });
const result = await client.generateImage({
dimensions: { width_in_px: 1080, height_in_px: 1080 },
output_format: "jpeg",
layers: [
{
index: 0,
type: "image",
file: {
type: "url",
name: "property.jpg",
url: "https://example.com/properties/412-ridgewood.jpg",
},
dimensions: { width_in_px: 1080, height_in_px: 720 },
},
{
index: 1,
type: "solid-color",
hex_color: "#1A1A2E",
position: { x_in_px: 0, y_in_px: 720 },
dimensions: { width_in_px: 1080, height_in_px: 360 },
},
{
index: 2,
type: "solid-color",
hex_color: "#E63946",
position: { x_in_px: 40, y_in_px: 40 },
dimensions: { width_in_px: 180, height_in_px: 48 },
border_radius: 6,
},
{
index: 3,
type: "text",
text: "JUST LISTED",
font_name: "Inter",
font_size_in_px: 18,
font_weight: "bold",
text_color: "#FFFFFF",
text_align: "center",
vertical_align: "center",
position: { x_in_px: 40, y_in_px: 40 },
dimensions: { width_in_px: 180, height_in_px: 48 },
},
{
index: 4,
type: "text",
text: "EUR 685,000",
font_name: "Inter",
font_size_in_px: 52,
font_weight: "bold",
text_color: "#FFFFFF",
text_align: "left",
vertical_align: "center",
position: { x_in_px: 40, y_in_px: 740 },
dimensions: { width_in_px: 700, height_in_px: 72 },
},
{
index: 5,
type: "text",
text: "412 Ridgewood Lane, Berlin 10115",
font_name: "Inter",
font_size_in_px: 22,
font_weight: "regular",
text_color: "#CBD5E1",
text_align: "left",
vertical_align: "center",
position: { x_in_px: 40, y_in_px: 820 },
dimensions: { width_in_px: 900, height_in_px: 36 },
},
{
index: 6,
type: "text",
text: "4 bed · 2.5 bath · 2,140 sq ft",
font_name: "Inter",
font_size_in_px: 20,
font_weight: "regular",
text_color: "#94A3B8",
text_align: "left",
vertical_align: "center",
position: { x_in_px: 40, y_in_px: 870 },
dimensions: { width_in_px: 900, height_in_px: 36 },
},
{
index: 7,
type: "text",
text: "Hanse Realty GmbH",
font_name: "Inter",
font_size_in_px: 18,
font_weight: "semibold",
text_color: "#64748B",
text_align: "right",
vertical_align: "center",
position: { x_in_px: 40, y_in_px: 1020 },
dimensions: { width_in_px: 1000, height_in_px: 36 },
},
],
});
from iterationlayer import IterationLayer
client = IterationLayer(api_key="YOUR_API_KEY")
result = client.generate_image(
dimensions={"width_in_px": 1080, "height_in_px": 1080},
output_format="jpeg",
layers=[
{
"index": 0,
"type": "image",
"file": {
"type": "url",
"name": "property.jpg",
"url": "https://example.com/properties/412-ridgewood.jpg",
},
"dimensions": {"width_in_px": 1080, "height_in_px": 720},
},
{
"index": 1,
"type": "solid-color",
"hex_color": "#1A1A2E",
"position": {"x_in_px": 0, "y_in_px": 720},
"dimensions": {"width_in_px": 1080, "height_in_px": 360},
},
{
"index": 2,
"type": "solid-color",
"hex_color": "#E63946",
"position": {"x_in_px": 40, "y_in_px": 40},
"dimensions": {"width_in_px": 180, "height_in_px": 48},
"border_radius": 6,
},
{
"index": 3,
"type": "text",
"text": "JUST LISTED",
"font_name": "Inter",
"font_size_in_px": 18,
"font_weight": "bold",
"text_color": "#FFFFFF",
"text_align": "center",
"vertical_align": "center",
"position": {"x_in_px": 40, "y_in_px": 40},
"dimensions": {"width_in_px": 180, "height_in_px": 48},
},
{
"index": 4,
"type": "text",
"text": "EUR 685,000",
"font_name": "Inter",
"font_size_in_px": 52,
"font_weight": "bold",
"text_color": "#FFFFFF",
"text_align": "left",
"vertical_align": "center",
"position": {"x_in_px": 40, "y_in_px": 740},
"dimensions": {"width_in_px": 700, "height_in_px": 72},
},
{
"index": 5,
"type": "text",
"text": "412 Ridgewood Lane, Berlin 10115",
"font_name": "Inter",
"font_size_in_px": 22,
"font_weight": "regular",
"text_color": "#CBD5E1",
"text_align": "left",
"vertical_align": "center",
"position": {"x_in_px": 40, "y_in_px": 820},
"dimensions": {"width_in_px": 900, "height_in_px": 36},
},
{
"index": 6,
"type": "text",
"text": "4 bed · 2.5 bath · 2,140 sq ft",
"font_name": "Inter",
"font_size_in_px": 20,
"font_weight": "regular",
"text_color": "#94A3B8",
"text_align": "left",
"vertical_align": "center",
"position": {"x_in_px": 40, "y_in_px": 870},
"dimensions": {"width_in_px": 900, "height_in_px": 36},
},
{
"index": 7,
"type": "text",
"text": "Hanse Realty GmbH",
"font_name": "Inter",
"font_size_in_px": 18,
"font_weight": "semibold",
"text_color": "#64748B",
"text_align": "right",
"vertical_align": "center",
"position": {"x_in_px": 40, "y_in_px": 1020},
"dimensions": {"width_in_px": 1000, "height_in_px": 36},
},
],
)
package main
import il "github.com/iterationlayer/sdk-go"
func main() {
client := il.NewClient("YOUR_API_KEY")
result, err := client.GenerateImage(il.GenerateImageRequest{
Dimensions: il.Dimensions{WidthInPx: 1080, HeightInPx: 1080},
OutputFormat: "jpeg",
Layers: []il.Layer{
il.NewImageLayer(
0,
il.FileInput{Type: "url", Name: "property.jpg", Url: "https://example.com/properties/412-ridgewood.jpg"},
il.Position{XInPx: 0, YInPx: 0},
il.Dimensions{WidthInPx: 1080, HeightInPx: 720},
),
il.NewSolidColorLayer(1, "#1A1A2E",
il.Position{XInPx: 0, YInPx: 720},
il.Dimensions{WidthInPx: 1080, HeightInPx: 360},
),
il.NewSolidColorLayer(2, "#E63946",
il.Position{XInPx: 40, YInPx: 40},
il.Dimensions{WidthInPx: 180, HeightInPx: 48},
),
il.NewTextLayer(3, "JUST LISTED", "Inter", 18, "#FFFFFF",
il.Position{XInPx: 40, YInPx: 40},
il.Dimensions{WidthInPx: 180, HeightInPx: 48},
),
il.NewTextLayer(4, "EUR 685,000", "Inter", 52, "#FFFFFF",
il.Position{XInPx: 40, YInPx: 740},
il.Dimensions{WidthInPx: 700, HeightInPx: 72},
),
il.NewTextLayer(5, "412 Ridgewood Lane, Berlin 10115", "Inter", 22, "#CBD5E1",
il.Position{XInPx: 40, YInPx: 820},
il.Dimensions{WidthInPx: 900, HeightInPx: 36},
),
il.NewTextLayer(6, "4 bed · 2.5 bath · 2,140 sq ft", "Inter", 20, "#94A3B8",
il.Position{XInPx: 40, YInPx: 870},
il.Dimensions{WidthInPx: 900, HeightInPx: 36},
),
il.NewTextLayer(7, "Hanse Realty GmbH", "Inter", 18, "#64748B",
il.Position{XInPx: 40, YInPx: 1020},
il.Dimensions{WidthInPx: 1000, HeightInPx: 36},
),
},
})
if err != nil {
panic(err)
}
_ = result
}
{
"name": "Generate Real Estate Listing Graphic",
"nodes": [
{
"parameters": {
"content": "## Generate Real Estate Listing Graphic\n\nReal estate agencies and property platforms use this recipe to generate listing graphics for social media, email campaigns, and digital signage. Feed in property data and a photo URL to get a print-ready branded card \u2014 no design software, no manual exports.\n\n**Note:** This workflow uses the Iteration Layer community node (`n8n-nodes-iterationlayer`). Install it via Settings > Community Nodes on self-hosted n8n, or add it directly on n8n Cloud with Verified Community Nodes enabled.",
"height_in_px": 280,
"width_in_px": 500,
"color": 2
},
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
200,
40
],
"id": "4c51f576-8de8-4023-b28e-4798e54af8d7",
"name": "Overview"
},
{
"parameters": {
"content": "### Step 1: Generate Image\nResource: **Image Generation**\n\nConfigure the Image Generation parameters below, then connect your credentials.",
"height_in_px": 160,
"width_in_px": 300,
"color": 6
},
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
475,
100
],
"id": "43d2927f-bae6-4ae8-aef7-177d92f1aac5",
"name": "Step 1 Note"
},
{
"parameters": {},
"type": "n8n-nodes-base.manualTrigger",
"typeVersion": 1,
"position": [
250,
300
],
"id": "d0e1f2a3-aaaa-4000-8000-000000000001",
"name": "Manual Trigger"
},
{
"parameters": {
"resource": "imageGeneration",
"widthInPx": 1080,
"heightInPx": 1080,
"outputFormat": "jpeg",
"layersJson": "[{\"index\":0,\"type\":\"image\",\"file\":{\"type\":\"url\",\"name\":\"property.jpg\",\"url\":\"https://example.com/properties/412-ridgewood.jpg\"},\"dimensions\":{\"width\":1080,\"height\":720}},{\"index\":1,\"type\":\"solid-color\",\"hex_color\":\"#1A1A2E\",\"position\":{\"x\":0,\"y\":720},\"dimensions\":{\"width\":1080,\"height\":360}},{\"index\":2,\"type\":\"solid-color\",\"hex_color\":\"#E63946\",\"position\":{\"x\":40,\"y\":40},\"dimensions\":{\"width\":180,\"height\":48},\"border_radius\":6},{\"index\":3,\"type\":\"text\",\"text\":\"JUST LISTED\",\"font_name\":\"Inter\",\"font_size_in_px\":18,\"font_weight\":\"bold\",\"text_color\":\"#FFFFFF\",\"text_align\":\"center\",\"vertical_align\":\"center\",\"position\":{\"x\":40,\"y\":40},\"dimensions\":{\"width\":180,\"height\":48}},{\"index\":4,\"type\":\"text\",\"text\":\"EUR 685,000\",\"font_name\":\"Inter\",\"font_size_in_px\":52,\"font_weight\":\"bold\",\"text_color\":\"#FFFFFF\",\"text_align\":\"left\",\"vertical_align\":\"center\",\"position\":{\"x\":40,\"y\":740},\"dimensions\":{\"width\":700,\"height\":72}},{\"index\":5,\"type\":\"text\",\"text\":\"412 Ridgewood Lane, Berlin 10115\",\"font_name\":\"Inter\",\"font_size_in_px\":22,\"font_weight\":\"regular\",\"text_color\":\"#CBD5E1\",\"text_align\":\"left\",\"vertical_align\":\"center\",\"position\":{\"x\":40,\"y\":820},\"dimensions\":{\"width\":900,\"height\":36}},{\"index\":6,\"type\":\"text\",\"text\":\"4 bed \\u00b7 2.5 bath \\u00b7 2,140 sq ft\",\"font_name\":\"Inter\",\"font_size_in_px\":20,\"font_weight\":\"regular\",\"text_color\":\"#94A3B8\",\"text_align\":\"left\",\"vertical_align\":\"center\",\"position\":{\"x\":40,\"y\":870},\"dimensions\":{\"width\":900,\"height\":36}},{\"index\":7,\"type\":\"text\",\"text\":\"Hanse Realty GmbH\",\"font_name\":\"Inter\",\"font_size_in_px\":18,\"font_weight\":\"semibold\",\"text_color\":\"#64748B\",\"text_align\":\"right\",\"vertical_align\":\"center\",\"position\":{\"x\":40,\"y\":1020},\"dimensions\":{\"width\":1000,\"height\":36}}]",
"fontsJson": "[]"
},
"type": "n8n-nodes-iterationlayer.iterationLayer",
"typeVersion": 1,
"position": [
500,
300
],
"id": "d0e1f2a3-aaaa-4000-8000-000000000002",
"name": "Generate Image",
"credentials": {
"iterationLayerApi": {
"id": "1",
"name": "Iteration Layer API"
}
}
}
],
"connections": {
"Manual Trigger": {
"main": [
[
{
"node": "Generate Image",
"type": "main",
"index": 0
}
]
]
}
},
"settings": {
"executionOrder": "v1"
}
}
Generate a real estate listing graphic at 1200x628 pixels. Use the generate_image tool with these layers:
1. Property photo: image layer with [property image URL] as full background
2. Gradient overlay: gradient layer for text readability at the bottom
3. Status badge: solid-color rectangle with [status] text (e.g., "For Sale", "Sold")
4. Price: text layer with [price] in large bold text
5. Address: text layer with [street address, city, state]
6. Stats: text layers for [bedrooms] bd, [bathrooms] ba, [square feet] sqft
7. Agency logo: image layer with [logo URL]
{
"success": true,
"data": {
"buffer": "/9j/4AAQSkZJRgABAQ...",
"mime_type": "image/jpeg"
}
}
아직 리뷰가 없습니다. 사용해 본 소감을 남겨 주세요.
로그인 없이 쓸 수 있습니다. 사용 후기, 질문, 작동하지 않은 환경 등을 편하게 남겨 주세요.
아직 댓글이 없습니다. 첫 댓글을 남겨 보세요.
Resize, sharpen, and compress an image to fit email platform size limits in a single pipeline.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Compress an image to fit within a specific file size in bytes using quality-first compression.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Convert a contract PDF to clean markdown for clause extraction or LLM analysis.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Convert external documents — specs, contracts, reports — to markdown for knowledge base ingestion.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Convert a document to clean markdown suitable for chunking and embedding in a RAG pipeline.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Convert an image between PNG, JPEG, and WebP formats with quality control for web optimization.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.