Turn ideas into movies with DeepSeek, RunwayML, ElevenLabs & Creatomate
How It Works Story Generation – Your idea is transformed into a narrative split into scenes using DeepSeek LLM. Visuals – Each scene is illustrated with AI images via Replicate, then animated into cinematic video clips with RunwayML. Voice & Music – Narration is created using ElevenLabs (text-to-speech), while Replicate audio models generate background music. Final Assembly – All assets are merged into a professional video using Creatomate. Delivery – Everything is orchestrated by n8n, triggered from Slack with /render, and the final video link is delivered back instantly. --- Workflow in Action Trigger from Slack Type your idea with /render in Slack - the workflow starts automatically. Final Video Output Receive a polished cinematic video link in Slack. Creatomate Template ⚠️ Important: You must create your own template in Creatomate. This is a one-time setup - the template defines where the voiceover, music, and video clips will be placed. The more detailed and refined your template is, the better the final cinematic result. --- Required APIs To run this workflow, you need accounts and API keys from the following services: DeepSeek – Story generation (LLM) Replicate – Images & AI music generation RunwayML – Image-to-video animations ElevenLabs – Text-to-speech voiceovers Creatomate – Video rendering and templates Dropbox – File storage and asset syncing Slack – Workflow trigger and video delivery --- Setup Steps Import the JSON workflow into your n8n instance. Add your API credentials for each service above. Create a Creatomate template (only once) – define layers for visuals, voice, and music. Trigger the workflow from Slack with /render Your Story Idea. Receive your final cinematic video link directly in Slack. --- Use Cases Automated YouTube Shorts / TikToks for faceless content creators. Scalable ad creatives and marketing videos for agencies. Educational explainers and onboarding videos generated from text. Rapid prototyping of cinematic ideas for developers & storytellers. --- With this workflow, you’re not just using AI tools – you’re running a full AI-powered studio in n8n.
By Țugui DragoșDynamic MCP server selection with OpenAI GPT-4.1 and contextual AI reranker
PROBLEM Thousands of MCP Servers exist and many are updated daily, making server selection difficult for LLMs. Current approaches require manually downloading and configuring servers, limiting flexibility. When multiple servers are pre-configured, LLMs get overwhelmed and confused about which server to use for specific tasks. This template enables dynamic server selection from a live PulseMCP directory of 5000+ servers. How it works A user query goes to an LLM that decides whether to use MCP servers to fulfill a given query and provides reasoning for its decision. Next, we fetch MCP Servers from Pulse MCP API and format them as documents for reranking Now, we use Contextual AI's Reranker to score and rank all MCP Servers based on our query and instructions How to set up Sign up for a free trial of Contextual AI here to find CONTEXTUALAIAPIKEY. Click on variables option in left panel and add a new environment variable CONTEXTUALAIAPIKEY. For the baseline model, we have used GPT 4.1 mini, you can find your OpenAI API key here How to customize the workflow We use chat trigger to initate the workflow. Feel free to replace it with a webhook or other trigger as required. We use OpenAI's GPT 4.1 mini as the baseline model and reranker prompt generator. You can swap out this section to use the LLM of your choice. We fetch 5000 MCP Servers from the PulseMCP directory as a baseline number, feel free to adjust this parameter as required. We are using Contextual AI's ctxl-rerank-v2-instruct-multilingual reranker model, which can be swapped with any one of the following rerankers: 1) ctxl-rerank-v2-instruct-multilingual 2) ctxl-rerank-v2-instruct-multilingual-mini 3) ctxl-rerank-v1-instruct You can checkout this blog for more information about rerankers to learn more about them. Good to know: Contextual AI Reranker (with full MCP docs): ~$0.035/query Includes 0.035 for reranking + ~$0.0001 for OpenAI instruction generation. OpenAI Baseline: ~$0.017/query
By Jinash RouniyarAutomated blog creation from news to publication using Gemini, Ideogram & Slack
The AI Blog Creator with Gemini, Replicate Image, Supabase Publishing & Slack is a fully automated content generation and publishing workflow designed for modern marketing and SaaS teams. It automatically fetches the latest industry trends, generates SEO-optimized blogs using AI, creates a relevant featured image, publishes the post to your CMS (e.g., Supabase or custom API), and notifies your team via Slack all on a daily schedule. This workflow connects multiple services NewsAPI, Google Gemini, Replicate, Supabase, and Slack into one intelligent content pipeline that runs hands-free once set up. --- ✨ Features 📰 Fetch Trending Topics — pulls the latest news or updates from your selected industry (via NewsAPI). 🤖 AI Topic Generation — Gemini suggests trending blog topics relevant to AI, SaaS, and Automation. 📝 AI Blog Authoring — Gemini then writes a full 1200-1500 word SEO-optimized article in Markdown. 🧹 Smart JSON Cleaner — A resilient code node parses Gemini’s output and ensures clean, structured data. 🖼️ Auto-Generated Image — Replicate’s Ideogram model creates a blog cover image based on the content prompt. 🌐 Automatic Publishing — Posts are automatically published to your Supabase or custom backend. 💬 Slack Notification — Notifies your team with blog details and live URL. ⏰ Fully Scheduled — Runs automatically every day at your preferred time (default 10 AM IST). --- ⚙️ Workflow Structure | Step | Node | Purpose | | ---- | ----------------------------------- | ----------------------------------------------- | | 1 | Schedule Trigger | Runs daily at 10 AM | | 2 | Fetch Industry Trends (NewsAPI) | Retrieves trending articles | | 3 | Message a model (Gemini) | Generates trending topic ideas | | 4 | Message a model1 (Gemini) | Writes full SEO blog content | | 5 | Code in JavaScript | Cleans, validates, and normalizes Gemini output | | 6 | HTTP Request (Replicate) | Generates an image using Ideogram | | 7 | HTTP Request1 | Retrieves generated image URL | | 8 | Wait + If | Polls until image generation succeeds | | 9 | Edit Fields | Assembles blog fields into final JSON | | 10 | Publish to Supabase | Posts to your CMS | | 11 | Slack Notification | Sends message to your Slack channel | --- 🔧 Setup Instructions Import the Workflow in n8n and enable it. Create the following credentials: NewsAPI (Query Auth) — from https://newsapi.org Google Gemini (PaLM API) — use your Gemini API key Replicate (Bearer Auth) — API key from https://replicate.com/account Supabase (Header Auth) — endpoint to your /functions/v1/blog-api (set your key in header) Slack API — create a Slack App token with chat:write permission Edit the NewsAPI URL query parameter to match your industry (e.g., q=AI automation SaaS). Update the Supabase publish URL to your project endpoint if needed. Adjust the Slack Channel name under “Slack Notification”. (Optional) Change the Schedule Trigger time as per your timezone. --- 💡 Notes & Tips The Code in JavaScript node is robust against malformed or extra text in Gemini output — it sanitizes Markdown and reconstructs clean JSON safely. You can replace Supabase with any CMS or Webhook endpoint by editing the “Publish to Supabase” node. The Replicate model used is ideogram-ai/ideogram-v3-turbo — you can swap it with Stable Diffusion or another model for different aesthetics. Use the slug field in your blog URLs for SEO-friendly links. Test with one manual execution before activating scheduled runs. If Slack notification fails, verify the token scopes and channel permissions. --- 🧩 Tags AI Automation ContentMarketing BlogGenerator n8n Supabase Gemini Replicate Slack WorkflowAutomation
By Parag Javale