Fashion brand content calendar from mood boards to scheduled posts

Fashion marketing teams lose roughly eight hours each week moving mood board images between tools, writing captions manually, and scheduling posts one platform at a time. A designer creates a concept, a copywriter drafts the caption, someone uploads to a scheduling tool, and another person double-checks the timing. By the time the post goes live, the original creative momentum has long since faded. You can compress that entire sequence into a single workflow that runs in minutes instead of days. The secret is connecting three specialist tools with an orchestration layer that understands when and how to pass data between them. No manual downloads, no copy-paste, no switching tabs. This guide shows you how to build a fashion content calendar system that transforms mood board concepts into scheduled social posts automatically. You'll move from idea to publication without human intervention between steps.

The Automated Workflow

We'll use n8n as the orchestration engine because it handles image processing well and connects to all three tools without requiring external webhooks. The workflow moves from creative generation through to scheduled publication.

Step 1: Generate mood board images

Start by using Nsketch AI's image generation API to produce concept images based on your brand brief. You'll trigger this through n8n on a schedule or manually.

CopyPOST https://api.nsketch.ai/v1/generate
Content-Type: application/json
Authorization: Bearer YOUR_NSKETCH_API_KEY { "prompt": "minimalist autumn fashion mood board, earth tones, natural lighting, high fashion editorial", "model": "flux_pro", "num_images": 4, "style_template": "fashion_editorial", "width": 1080, "height": 1350
}

This returns four images ready for use. N8n stores these in memory and passes them forward to the next step.

Step 2: Refine with PickOrCraft for brand consistency

Rather than accepting generic generated images, you'll want to train PickOrCraft with your brand's aesthetic so subsequent generations match your house style. Set up your model once, then reference it by ID in the workflow.

CopyPOST https://api.pickorcraft.com/v1/generate
Content-Type: application/json
Authorization: Bearer YOUR_PICKORCRAFT_API_KEY { "model_id": "your_brand_model_12345", "prompts": [ "autumn fashion campaign featuring trained brand aesthetic", "minimalist product photography with earth tones", "editorial styling matching brand lookbook" ], "num_variations": 3, "output_format": "jpg"
}

The returned images now carry your brand's visual signature. N8n batches these with metadata: caption hooks, colour palettes, target audience, post type (feed, reel, story).

Step 3: Generate captions and metadata

Use Claude Opus 4.6 through the Anthropic API to write captions tailored to each image. Send the image URL and brand guidelines.

CopyPOST https://api.anthropic.com/v1/messages
Content-Type: application/json
Authorization: x-api-key YOUR_ANTHROPIC_API_KEY { "model": "claude-opus-4.6", "max_tokens": 500, "messages": [ { "role": "user", "content": [ { "type": "image", "source": { "type": "url", "url": "https://your-cdn.com/mood-board-1.jpg" } }, { "type": "text", "text": "You are a luxury fashion copywriter for an upmarket sustainable brand. Write a caption for this mood board image. Include a relevant hashtag strategy. Keep to 150 words. Format as JSON with keys: 'caption', 'hashtags', 'cta', 'best_posting_time'." } ] } ]
}

Parse the JSON response and store it alongside the image reference.

Step 4: Schedule posts to social media

Postwise handles the final step. The workflow formats each image and caption pair into Postwise's batch scheduling endpoint, staggering posts across your calendar.

CopyPOST https://api.postwise.com/v1/batch-schedule
Content-Type: application/json
Authorization: Bearer YOUR_POSTWISE_API_KEY { "posts": [ { "content": "Autumn edit now live. Sustainably crafted pieces for the season ahead. Explore the collection.", "image_url": "https://your-cdn.com/mood-board-1.jpg", "platforms": ["instagram", "twitter"], "scheduled_time": "2026-03-15T10:00:00Z", "hashtags": ["#sustainablefashion", "#autumnstyle", "#luxurylounge"] }, { "content": "Minimalism meets purpose. Every stitch tells a story.", "image_url": "https://your-cdn.com/mood-board-2.jpg", "platforms": ["instagram"], "scheduled_time": "2026-03-16T14:30:00Z", "hashtags": ["#slowfashion", "#ethicalstyle"] } ]
}

The entire sequence runs without you touching a file manager or social platform dashboard.

Putting it together in n8n

Create an n8n workflow with these nodes: 1. Manual trigger or cron schedule (runs daily at 9am) 2. HTTP POST to Nsketch AI for initial mood board generation 3. Loop through returned images 4. HTTP POST to PickOrCraft with your trained model ID 5. For each refined image, send to Claude Opus 4.6 via HTTP POST 6. Parse JSON caption response, build metadata object 7. HTTP POST batch to Postwise with all formatted posts 8. Log completion and send Slack notification with preview of scheduled posts Between nodes 3 and 4, add error handling: if PickOrCraft fails, fall back to Nsketch output rather than stalling the workflow. Between nodes 5 and 6, validate that Claude returned valid JSON; if not, regenerate with a simpler prompt.

The Manual Alternative

If you prefer hands-on control over creative output, skip the n8n automation and use these tools sequentially without workflow integration. Generate your mood board in Nsketch AI directly through their web interface. Download the outputs, select your favourite variants, and upload them to PickOrCraft's dashboard to refine against your trained brand model. Write captions manually or paste image URLs into Claude (via web or API) for generation. Then import final images and text into Postwise and schedule manually by platform and date. This approach lets you review and adjust at every stage but takes roughly six to eight hours per content calendar versus thirty minutes with the automated workflow.

Pro Tips Rate limit management:

Nsketch and PickOrCraft both enforce per-minute generation limits on their standard plans. Stagger API calls in your n8n workflow using the "Wait" node; space requests at least two seconds apart. This prevents 429 errors mid-run. Image storage: Don't rely on temporary URLs from generation APIs. After each step returns images, download them to your own CDN (Cloudinary, AWS S3) and reference those URLs in downstream steps. This keeps your posts live even if the original API cache expires. Caption variation: Send the same image to Claude three times with slightly different prompts (formal tone, conversational tone, question-driven) and let Postwise choose the best-performing version based on historical CTR. Store all three variants and rotate them. Cost optimisation: Use GPT-4o mini instead of Opus for caption generation if you're running this daily. The quality difference is minimal for social media copy, and you'll save roughly 70 percent on token costs. Reserve Opus for complex multi-image comparisons or brand guideline analysis. Failure recovery: If Postwise scheduling fails (platform downtime, auth token expired), n8n should queue posts to a fallback tool like Buffer or save them to a Google Sheet for manual review. Add a conditional node that detects HTTP 5xx responses and triggers backup logic.

Cost Breakdown