AI Prompt Library — Version-Controlled Creative Prompt Management

What It Does

This n8n workflow creates a searchable, version-controlled library of every AI prompt used in creative production — across Midjourney, Adobe Firefly, Stable Diffusion, and DALL-E. Log prompts, auto-tag them with GPT-4o, track iterations, and never lose a winning prompt again.

Key Features

• Prompt Logging — Captures the full prompt, negative prompt, AI tool, model version, parameters (seed, CFG, steps), and the generated output URL
• AI Auto-Tagging — GPT-4o analyzes each prompt and auto-tags it by style, mood, lighting, composition, materials, camera angle, color palette, and category
• Version Tracking — Links prompt iterations together so you can trace the evolution from first draft to approved asset
• Visual Thumbnails — Downloads and stores thumbnail previews alongside prompts for quick visual browsing
• Search & Recall — Find any prompt by tag, campaign, product, style, or natural language description
• Team Sharing — Approved prompts are published to a shared Slack channel so the whole team can reuse winning formulas

AI Tools Supported

• Midjourney — Prompt, parameters (–ar, –v, –s, –c), seed, job ID, image URL
• Adobe Firefly — Prompt, style presets, content type, visual intensity, reference images
• Stable Diffusion — Prompt, negative prompt, model, sampler, CFG scale, steps, seed
• DALL-E — Prompt, model version (DALL-E 3), size, quality, style setting

How It Works

Webhook receives prompt data and output URL → GPT-4o auto-tags the prompt by style, mood, lighting, composition, and category → Version links are established for prompt iterations → Thumbnail is downloaded → Full record saved to Google Sheets with all metadata → Team notified on Slack.

Tech Stack

n8n workflow automation, OpenAI GPT-4o (auto-tagging), Google Sheets API, Slack API, webhook-driven architecture.

View on GitHub →