The workflow JSON
Copy or download the full n8n JSON below. Paste it into a new n8n workflow, add your credentials, activate. Full import guide →
{
"name": "AgentForge \u2014 Jewellery AI Agent (FAL Nano Banana Edit) \u00b7 single + bulk",
"nodes": [
{
"parameters": {
"httpMethod": "POST",
"path": "generate-jewellery",
"responseMode": "responseNode",
"options": {}
},
"id": "webhook-jewellery",
"name": "Webhook \u00b7 /generate-jewellery",
"type": "n8n-nodes-base.webhook",
"typeVersion": 1.1,
"position": [
220,
360
]
},
{
"parameters": {
"jsCode": "// ============================================================\n// AGENTFORGE JEWELLERY AGENT \u2014 Normalize + Master prompt builder\n// Handles BOTH single and bulk modes.\n// Input : webhook body from app/api/jewellery/generate/route.ts\n// Output : N items (1 per image), each with master_prompt + source URL\n// ============================================================\n\nconst body = $json.body || $json;\nconst s = body.shared_settings || {};\nconst mode = body.generation_mode || 'single';\nconst batchId = body.batch_id || null;\n\nconst rawItems = Array.isArray(body.items) && body.items.length\n ? body.items\n : [{\n generation_id: body.generation_id,\n source_image_url: body.source_image_url || s.source_image_url,\n original_name: 'image-1'\n }];\n\nif (!rawItems.length) throw new Error('No items to process \u2014 payload has no images.');\n\n// ---------- Aspect / quality ----------\nconst sizeStr = String(s.output_size || 'Square 1080x1080');\nconst isMobile = /Mobile|1920|9:16/i.test(sizeStr);\nconst aspect = isMobile ? '9:16 vertical (1080x1920)' : '1:1 square (1080x1080)';\nconst quality = String(s.output_quality || 'Premium');\n\n// ---------- Branding reserve ----------\nconst cd = s.company_details || {};\nconst hasBrandingText = Boolean(cd.company_name) || Boolean(cd.website) || Boolean(cd.phone) || Boolean(cd.address);\nconst reserveBottomStrip = hasBrandingText;\nconst reserveTopRight = true;\n\n// ---------- Style guides ----------\nconst SHOOT_STYLE_GUIDE = {\n 'Luxury Studio': 'premium softbox studio lighting, dark velvet or gradient grey background, museum-quality jewellery showcase, dramatic key light on stones',\n 'White Catalogue': 'pure white seamless background, even shadowless ecommerce lighting, Amazon/Flipkart catalogue ready, ultra clean',\n 'Bridal Editorial': 'rich warm gold tones, ornate silk drape backdrop, candlelight + softbox mix, Vogue Bridal magazine feel',\n 'Macro Detail': 'extreme macro shot, ring-light reflections in every stone facet, shallow depth of field, gemstone-first composition',\n 'Lifestyle Campaign': 'natural daylight cafe / outdoor terrace mood, modern Indian woman lifestyle, soft bokeh, social-media ad feel',\n 'Indian Model': 'warm honey-tan Indian skin tone, natural Indian facial features, traditional or fusion styling, soft glow makeup',\n 'Bridal Look': 'heavy bridal styling with subtle mehndi, traditional bridal makeup, dupatta drape, regal posture, wedding photography feel',\n 'Luxury Editorial': 'high-fashion editorial aesthetic, dramatic shadow play, designer-magazine feel, model styled as luxury brand muse',\n 'Minimal Modern': 'minimal scandi-influenced styling, soft neutral palette, modern Indian woman, clean understated elegance'\n};\nconst MODEL_TYPE_GUIDE = {\n 'No Model': 'product-only \u2014 no human, no hand. Pure jewellery composition on the chosen surface/prop.',\n 'Hand Model': 'elegant feminine hand with manicured natural nails. Hand placement keeps the jewellery as the visual hero.',\n 'Couple Hands': 'two hands together (engagement / wedding feel) \u2014 one masculine, one feminine. Subtle warmth.',\n 'Female Model': 'Indian female adult model, jewellery clearly visible, styled around the piece.',\n 'Bridal Model': 'Indian bride in full bridal styling \u2014 dupatta, traditional outfit, bridal makeup, regal pose.',\n 'Luxury Flat Lay': 'top-down flat lay on a premium surface (velvet / silk / marble). Jewellery is the hero with curated props.',\n 'Macro Detail': 'extreme zoom on the jewellery only \u2014 no model, no full piece view. Focus on a single stone / facet / engraving.',\n 'Ear Close-up': 'extreme close-up of one ear wearing the earring, side profile, focus on earring sparkle.',\n 'Bust Portrait': 'shoulders-up portrait with face and jewellery both visible, elegant pose.',\n 'Half Body': 'waist-up lifestyle shot showing model styling around the jewellery.',\n 'Full Body': 'editorial full-body shot with the jewellery as the focal accessory.',\n 'Neck Focus': 'tight crop on neckline showing the necklace draped naturally.',\n 'Neck Close-up': 'extreme close-up of neckline \u2014 jewellery detail and skin texture only.',\n 'Wrist Close-up': 'tight wrist crop with bracelet/kada as hero.',\n 'Lifestyle Hand': 'natural hand pose in lifestyle context (holding coffee cup, on lap, etc.).',\n 'Detail Close-up': 'extreme detail crop of the jewellery \u2014 pattern, stone setting, engraving.',\n 'Editorial Scene': 'staged scene with model + jewellery in an editorial context (palace, garden, studio set).'\n};\nconst ACCESSORY_GUIDE = {\n 'No Accessories': 'no props \u2014 clean focus on the jewellery only.',\n 'Flat Lay': 'top-down flat lay arrangement on a premium fabric or surface.',\n 'Velvet Box': 'jewellery resting on or beside an open black/maroon velvet box \u2014 luxury showroom feel.',\n 'Marble Surface': 'placed on polished white-and-grey veined marble with subtle reflection.',\n 'Silk Drape': 'cascading silk fabric (cream, ivory or jewel-tone) as backdrop / under-layer.',\n 'Rose Petals': 'scattered fresh rose petals as romantic accent \u2014 sparse, tasteful.',\n 'Marigold + Diya': 'marigold flowers and a small lit diya \u2014 temple/festival jewellery context.',\n 'Diamond Sparkle Set': 'loose crystals and mirror shards catching light \u2014 adds sparkle drama.',\n 'Pearl String Decor': 'strands of pearls draped softly around the jewellery as accent.',\n 'Wooden Antique Tray': 'placed on a vintage carved wooden tray \u2014 heritage / antique jewellery context.'\n};\nconst CAMERA_ANGLE_GUIDE = {\n 'Auto Angle': 'choose the most flattering angle for this jewellery type',\n 'Eye Level': 'straight-on eye-level shot, parallel to subject',\n '45\u00b0 Angle': '45-degree tilt \u2014 classic jewellery showcase angle',\n 'Top Down': 'directly overhead flat-lay perspective',\n 'Side Profile': 'pure side view \u2014 best for earrings and pendants'\n};\nconst FACE_GUIDE = {\n 'Soft Smile': 'gentle warm smile, eyes relaxed',\n 'Confident': 'confident closed-mouth expression, direct gaze',\n 'Serious': 'serious editorial expression, eyes off-camera',\n 'Royal': 'regal bridal expression, chin slightly raised',\n 'Natural': 'candid natural expression, no posing'\n};\nconst pickGuide = (tbl, k, fb) => tbl[k] || `\"${k}\"${fb ? ' \u2014 ' + fb : ''}`;\n\nconst shootStyleDesc = pickGuide(SHOOT_STYLE_GUIDE, s.shoot_style, 'premium photography style');\nconst modelTypeDesc = pickGuide(MODEL_TYPE_GUIDE, s.model_type, 'as described');\nconst accessoryDesc = pickGuide(ACCESSORY_GUIDE, s.accessories, 'as described');\nconst cameraAngleDesc = pickGuide(CAMERA_ANGLE_GUIDE, s.camera_angle, 'as described');\nconst faceDesc = pickGuide(FACE_GUIDE, s.face_expression, 'natural expression');\n\nconst moreList = Array.isArray(s.more_jewellery) && s.more_jewellery.length\n ? ` (${s.more_jewellery.join(', ')})`\n : '';\nconst jewelLabel = `${s.jewellery_type || 'Jewellery piece'}${moreList}`;\n\nfunction buildPrompt(idx) {\n return `\nROLE: You are a senior luxury jewellery photographer + AI image editor working for an Indian D2C jewellery brand.\nOBJECTIVE: Take the reference jewellery photo I am providing and produce a single ${aspect} ${quality}-grade catalogue/editorial visual that strictly preserves the original jewellery while restyling the scene per the spec below.${mode === 'bulk' ? `\\n[Batch ${idx + 1} of ${rawItems.length} \u2014 keep style consistent across the batch.]` : ''}\n\n========================================================\n HARD RULES \u2014 NEVER VIOLATE\n========================================================\n1. PRESERVE THE JEWELLERY EXACTLY as in the reference image:\n - same overall SHAPE, contour and silhouette\n - same NUMBER of stones, same stone CUT, same stone COLOR\n - same METAL TONE (yellow gold / rose gold / silver / white gold / oxidized / kundan / polki)\n - same engravings, embossed patterns, polish texture, finishing\n - DO NOT redesign, restyle, recolor or \"improve\" the piece. Style the scene AROUND it.\n2. NO TEXT, NO SKU, NO PRODUCT CODE, NO BARCODE, NO NUMBERS, NO PRICE, NO LABELS anywhere on or near the jewellery. The image must be 100% free of typography and codes.\n3. NO LOGOS of any kind baked into the image \u2014 \"AgentForge\", \"AF\", brand watermarks, etc. are added separately client-side after this image is generated. Your output should be a clean photograph only.\n4. ${reserveBottomStrip ? 'RESERVE the BOTTOM 14% of the canvas as a clean, gently blurred, neutral surface. No jewellery, no model face, no busy props in that strip. A branding text strip will be overlaid post-process.' : 'Compose freely across the full canvas.'}\n5. ${reserveTopRight ? 'Keep the TOP-RIGHT corner (roughly 18% width x 18% height) visually calm \u2014 avoid bright highlights, faces or critical jewellery detail there. A small logo badge will be overlaid post-process.' : ''}\n6. Output must be a SINGLE photograph \u2014 no collages, no split frames, no before/after panels.\n\n========================================================\n SHOOT SPEC\n========================================================\n- Jewellery: ${jewelLabel}\n- Shoot Style & Model Look: \"${s.shoot_style}\" \u2192 ${shootStyleDesc}\n- Accessories / Props: \"${s.accessories}\" \u2192 ${accessoryDesc}\n- Model Type: \"${s.model_type}\" \u2192 ${modelTypeDesc}\n- Pose: \"${s.pose}\"\n- Face Expression: \"${s.face_expression}\" \u2192 ${faceDesc}\n- Camera Angle: \"${s.camera_angle}\" \u2192 ${cameraAngleDesc}\n- Output Frame: ${aspect}\n- Quality Tier: ${quality}${quality === 'Ultra HD' ? ' \u2014 push micro-detail, extra sharpness, crisper stone facets, finer skin texture' : ''}\n\n========================================================\n CLIENT NOTES (must be respected verbatim)\n========================================================\n- Jewellery protection notes: ${s.jewellery_notes ? `\"${s.jewellery_notes}\"` : '(none provided \u2014 default to faithful preservation)'}\n- Model styling notes: ${s.model_notes ? `\"${s.model_notes}\"` : '(none)'}\n- Custom instruction: ${s.custom_instruction ? `\"${s.custom_instruction}\"` : '(none)'}\n\n========================================================\n TECHNICAL\n========================================================\n- DSLR-grade detail, accurate metallic reflections, true-to-life gemstone color\n- Studio softbox + rim light unless the chosen style says otherwise\n- Skin tone for Indian models: warm honey-tan, natural Indian facial features, soft minimal makeup unless Bridal style is selected\n- Color grading: jewellery-first \u2014 background must support, never compete\n- Aspect: ${aspect}\n- Final: clean, share-ready, no text, no logos baked in, no SKU\n\nReturn ONE image only.\n`.trim();\n}\n\nconst out = rawItems.map((item, idx) => {\n if (!item.source_image_url) throw new Error(`Item ${idx} missing source_image_url`);\n return {\n json: {\n mode,\n batch_id: batchId,\n batch_size: rawItems.length,\n batch_index: idx,\n generation_id: item.generation_id,\n user_id: body.user_id,\n source_image_url: item.source_image_url,\n original_name: item.original_name || `image-${idx + 1}`,\n aspect_ratio: isMobile ? '9:16' : '1:1',\n is_free_account: Boolean(s.af_watermark),\n company_details: cd,\n master_prompt: buildPrompt(idx)\n }\n };\n});\nreturn out;\n"
},
"id": "normalize-and-prompt",
"name": "Normalize + Build Per-Item Prompts",
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
460,
360
]
},
{
"parameters": {
"batchSize": 1,
"options": {}
},
"id": "split-batches",
"name": "Split In Batches (1 / loop)",
"type": "n8n-nodes-base.splitInBatches",
"typeVersion": 3,
"position": [
700,
360
]
},
{
"parameters": {
"method": "POST",
"url": "https://fal.run/fal-ai/nano-banana/edit",
"sendHeaders": true,
"headerParameters": {
"parameters": [
{
"name": "Authorization",
"value": "=Key {{ $env.FAL_KEY }}"
},
{
"name": "Content-Type",
"value": "application/json"
}
]
},
"sendBody": true,
"specifyBody": "json",
"jsonBody": "={\n \"prompt\": {{ JSON.stringify($json.master_prompt) }},\n \"image_urls\": [{{ JSON.stringify($json.source_image_url) }}],\n \"num_images\": 1,\n \"output_format\": \"png\"\n}",
"options": {
"timeout": 180000
}
},
"id": "fal-nano-banana-edit",
"name": "FAL \u00b7 Nano Banana Edit (sync)",
"type": "n8n-nodes-base.httpRequest",
"typeVersion": 4.2,
"position": [
940,
360
],
"continueOnFail": true
},
{
"parameters": {
"jsCode": "// Read FAL sync response \u2192 pull the first image URL.\n// FAL sync returns: { images: [{ url, content_type, width, height }], description, seed, ... }\nconst resp = $json;\nconst meta = $('Normalize + Build Per-Item Prompts').item.json;\n\n// HTTP node error path (continueOnFail) may give us { error: ... }\nif (resp?.error || resp?.detail) {\n return [{\n json: {\n ...meta,\n ok: false,\n error_message: 'FAL error: ' + JSON.stringify(resp.error || resp.detail).slice(0, 400)\n }\n }];\n}\n\nconst images = resp?.images || resp?.data?.images || [];\nconst first = images[0];\nif (!first?.url) {\n return [{\n json: {\n ...meta,\n ok: false,\n error_message: 'FAL returned no image. Raw: ' + JSON.stringify(resp).slice(0, 400)\n }\n }];\n}\n\nconst mime = first.content_type || 'image/png';\nconst ext = mime.includes('jpeg') ? 'jpg' : 'png';\nconst objectPath = `jewellery-outputs/${meta.user_id || 'guest'}/${Date.now()}-${meta.generation_id}.${ext}`;\n\nreturn [{\n json: {\n ...meta,\n ok: true,\n fal_image_url: first.url,\n output_mime: mime,\n output_ext: ext,\n object_path: objectPath,\n fal_seed: resp.seed || null\n }\n}];\n"
},
"id": "extract-fal-result",
"name": "Extract FAL Result",
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
1180,
360
]
},
{
"parameters": {
"conditions": {
"options": {
"caseSensitive": true,
"leftValue": "",
"typeValidation": "loose"
},
"conditions": [
{
"id": "ok-check",
"leftValue": "={{ $json.ok }}",
"rightValue": true,
"operator": {
"type": "boolean",
"operation": "true",
"singleValue": true
}
}
],
"combinator": "and"
},
"options": {}
},
"id": "if-success",
"name": "IF \u00b7 success?",
"type": "n8n-nodes-base.if",
"typeVersion": 2,
"position": [
1420,
360
]
},
{
"parameters": {
"url": "={{ $json.fal_image_url }}",
"options": {
"response": {
"response": {
"responseFormat": "file"
}
},
"timeout": 60000
}
},
"id": "download-fal-result",
"name": "Download FAL Result Image",
"type": "n8n-nodes-base.httpRequest",
"typeVersion": 4.2,
"position": [
1660,
240
]
},
{
"parameters": {
"method": "POST",
"url": "={{ $env.SUPABASE_URL }}/storage/v1/object/designs/{{ $json.object_path }}",
"sendHeaders": true,
"headerParameters": {
"parameters": [
{
"name": "Authorization",
"value": "=Bearer {{ $env.SUPABASE_SERVICE_ROLE_KEY }}"
},
{
"name": "apikey",
"value": "={{ $env.SUPABASE_SERVICE_ROLE_KEY }}"
},
{
"name": "Content-Type",
"value": "={{ $json.output_mime }}"
},
{
"name": "x-upsert",
"value": "true"
},
{
"name": "Cache-Control",
"value": "3600"
}
]
},
"sendBody": true,
"contentType": "binaryData",
"inputDataFieldName": "data",
"options": {
"timeout": 60000
}
},
"id": "upload-supabase",
"name": "Upload to Supabase Storage",
"type": "n8n-nodes-base.httpRequest",
"typeVersion": 4.2,
"position": [
1900,
240
]
},
{
"parameters": {
"jsCode": "// Build the public URL for the freshly uploaded Supabase object.\nconst meta = $('Extract FAL Result').item.json;\nconst supabaseUrl = $env.SUPABASE_URL;\nconst publicUrl = `${supabaseUrl}/storage/v1/object/public/designs/${meta.object_path}`;\nreturn [{ json: { ...meta, output_url: publicUrl } }];\n"
},
"id": "build-public-url",
"name": "Build Public URL",
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
2140,
240
]
},
{
"parameters": {
"method": "PATCH",
"url": "={{ $env.SUPABASE_URL }}/rest/v1/generations",
"sendQuery": true,
"queryParameters": {
"parameters": [
{
"name": "id",
"value": "=eq.{{ $json.generation_id }}"
}
]
},
"sendHeaders": true,
"headerParameters": {
"parameters": [
{
"name": "Authorization",
"value": "=Bearer {{ $env.SUPABASE_SERVICE_ROLE_KEY }}"
},
{
"name": "apikey",
"value": "={{ $env.SUPABASE_SERVICE_ROLE_KEY }}"
},
{
"name": "Content-Type",
"value": "application/json"
},
{
"name": "Prefer",
"value": "return=minimal"
}
]
},
"sendBody": true,
"specifyBody": "json",
"jsonBody": "={\n \"status\": \"completed\",\n \"output_url\": {{ JSON.stringify($json.output_url) }},\n \"updated_at\": {{ JSON.stringify(new Date().toISOString()) }}\n}"
},
"id": "update-generations",
"name": "Update generations row (completed)",
"type": "n8n-nodes-base.httpRequest",
"typeVersion": 4.2,
"position": [
2380,
240
]
},
{
"parameters": {
"method": "PATCH",
"url": "={{ $env.SUPABASE_URL }}/rest/v1/generations",
"sendQuery": true,
"queryParameters": {
"parameters": [
{
"name": "id",
"value": "=eq.{{ $json.generation_id }}"
}
]
},
"sendHeaders": true,
"headerParameters": {
"parameters": [
{
"name": "Authorization",
"value": "=Bearer {{ $env.SUPABASE_SERVICE_ROLE_KEY }}"
},
{
"name": "apikey",
"value": "={{ $env.SUPABASE_SERVICE_ROLE_KEY }}"
},
{
"name": "Content-Type",
"value": "application/json"
},
{
"name": "Prefer",
"value": "return=minimal"
}
]
},
"sendBody": true,
"specifyBody": "json",
"jsonBody": "={\n \"status\": \"failed\",\n \"error_message\": {{ JSON.stringify($json.error_message || 'Unknown error') }},\n \"updated_at\": {{ JSON.stringify(new Date().toISOString()) }}\n}"
},
"id": "mark-failed",
"name": "Mark Failed",
"type": "n8n-nodes-base.httpRequest",
"typeVersion": 4.2,
"position": [
1660,
480
]
},
{
"parameters": {
"jsCode": "// Collect this loop iteration's result into workflow static data.\nconst meta = $json;\nconst state = $getWorkflowStaticData('node');\nstate.results = state.results || [];\nstate.results.push({\n generation_id: meta.generation_id,\n status: meta.ok === false ? 'failed' : 'completed',\n output_url: meta.output_url || null,\n error_message: meta.error_message || null,\n original_name: meta.original_name\n});\nreturn [{ json: meta }];\n"
},
"id": "collect-results",
"name": "Collect Result",
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
2620,
360
]
},
{
"parameters": {
"jsCode": "// Runs after Split In Batches finishes all iterations.\n// Reads per-item results from workflow static data and emits a single item\n// containing the final response object \u2014 'Respond to Webhook' then serialises\n// it via {{ $json }}.\nconst state = $getWorkflowStaticData('node');\nconst results = Array.isArray(state.results) ? state.results : [];\nstate.results = []; // reset for next webhook run\n\nconst first = $('Normalize + Build Per-Item Prompts').first().json || {};\nconst completed = results.filter(r => r.status === 'completed');\nconst failed = results.filter(r => r.status === 'failed');\n\nreturn [{\n json: {\n success: failed.length === 0,\n mode: first.mode || 'single',\n batch_id: first.batch_id || null,\n total: results.length,\n completed: completed.length,\n failed: failed.length,\n // single-mode convenience fields\n generation_id: results[0]?.generation_id || null,\n image_url: results[0]?.output_url || null,\n output_url: results[0]?.output_url || null,\n // full per-item map (bulk mode)\n results\n }\n}];\n"
},
"id": "build-response-body",
"name": "Build Response Body",
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
2860,
360
]
},
{
"parameters": {
"respondWith": "json",
"responseBody": "={{ $json }}",
"options": {}
},
"id": "respond-webhook",
"name": "Respond to Webhook",
"type": "n8n-nodes-base.respondToWebhook",
"typeVersion": 1.1,
"position": [
3100,
360
]
}
],
"connections": {
"Webhook \u00b7 /generate-jewellery": {
"main": [
[
{
"node": "Normalize + Build Per-Item Prompts",
"type": "main",
"index": 0
}
]
]
},
"Normalize + Build Per-Item Prompts": {
"main": [
[
{
"node": "Split In Batches (1 / loop)",
"type": "main",
"index": 0
}
]
]
},
"Split In Batches (1 / loop)": {
"main": [
[
{
"node": "Build Response Body",
"type": "main",
"index": 0
}
],
[
{
"node": "FAL \u00b7 Nano Banana Edit (sync)",
"type": "main",
"index": 0
}
]
]
},
"FAL \u00b7 Nano Banana Edit (sync)": {
"main": [
[
{
"node": "Extract FAL Result",
"type": "main",
"index": 0
}
]
]
},
"Extract FAL Result": {
"main": [
[
{
"node": "IF \u00b7 success?",
"type": "main",
"index": 0
}
]
]
},
"IF \u00b7 success?": {
"main": [
[
{
"node": "Download FAL Result Image",
"type": "main",
"index": 0
}
],
[
{
"node": "Mark Failed",
"type": "main",
"index": 0
}
]
]
},
"Download FAL Result Image": {
"main": [
[
{
"node": "Upload to Supabase Storage",
"type": "main",
"index": 0
}
]
]
},
"Upload to Supabase Storage": {
"main": [
[
{
"node": "Build Public URL",
"type": "main",
"index": 0
}
]
]
},
"Build Public URL": {
"main": [
[
{
"node": "Update generations row (completed)",
"type": "main",
"index": 0
}
]
]
},
"Update generations row (completed)": {
"main": [
[
{
"node": "Collect Result",
"type": "main",
"index": 0
}
]
]
},
"Mark Failed": {
"main": [
[
{
"node": "Collect Result",
"type": "main",
"index": 0
}
]
]
},
"Collect Result": {
"main": [
[
{
"node": "Split In Batches (1 / loop)",
"type": "main",
"index": 0
}
]
]
},
"Build Response Body": {
"main": [
[
{
"node": "Respond to Webhook",
"type": "main",
"index": 0
}
]
]
}
},
"settings": {
"executionOrder": "v1",
"saveExecutionProgress": true,
"saveDataErrorExecution": "all",
"saveDataSuccessExecution": "all",
"errorWorkflow": ""
},
"tags": [
{
"name": "agentforge"
},
{
"name": "jewellery-ai"
}
]
}
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
AgentForge — Jewellery AI Agent (FAL Nano Banana Edit) · single + bulk. Uses httpRequest. Webhook trigger; 14 nodes.
Source: https://github.com/agentforgeindia/aiagentforge/blob/45c7b21a6f7502f1a23418ab02e7289e0728ff0b/n8n-workflows/jewellery-agent.json — original creator credit. Request a take-down →
Related workflows
Workflows that share integrations, category, or trigger type with this one. All free to copy and import.
Jigsaw API key for image processing, I use this as a gatekeeper/second pair of eyes. LINK to their website https://jigsawstack.com/ SECOND A postgress DATABASE (I use Supabase) LlamaCloud for the pars
HTTP Request, Postgres, Stop And Error +2
Onsite Photos to Jobs (SMS Agent). Uses dataTable, twilio, httpRequest, airtable. Webhook trigger; 62 nodes.
Data Table, Twilio, HTTP Request +1
W1 - IN WhatsApp Adapter (Secure + Fast ACK). Uses postgres, redis, httpRequest. Webhook trigger; 50 nodes.
Postgres, Redis, HTTP Request
W1 - IN WhatsApp Adapter (Secure + Fast ACK). Uses postgres, redis, httpRequest. Webhook trigger; 48 nodes.
Postgres, Redis, HTTP Request