Hey everyone 👋 Just wrapped up a project I've been building for the past few weeks and wanted to share it here since this community will appreciate the technical side. I built a modular 5-agent AI system for restaurant automation using n8n. Here's the architecture: 🔧 Tech stack: → n8n v2.6 (self-hosted + ngrok) → Supabase (PostgreSQL) — central database → OpenAI GPT-4o-mini + GPT-4o + DALL-E 3 → SerpAPI (Google Maps, TripAdvisor, Yelp scraping) → OpenWeatherMap API → Telegram Bot + Gmail OAuth2 → HTML/JS dashboard on Netlify ⚙️ The 5 agents: 1. Reputation Manager — scrapes reviews from 3 platforms, runs sentiment analysis, upserts to Supabase 2. Reservations Bot — multi-step Telegram conversation flow with session management in Supabase 3. Weather Menu Promoter — fetches weather, matches menu items by weather_tags, sends Telegram recommendations 4. Inventory Manager — reads stock, builds AI prompt, parses JSON response, sends low-stock alerts 5. AI Menu Photographer — dual mode: GPT-4o Vision analysis OR DALL-E 3 generation, stores in Supabase Storage A few things I learned building this: - Merge node must use Append mode (not Choose Branch) to avoid blocking - Telegram Parse Mode should be removed to prevent MarkdownV2 errors - OpenAI response lives at response.output[0].content[0].text Full workflows on GitHub (sanitized, no credentials): github.com/Evgeniy1970-ai/restaurant-ai-suite Happy to answer questions or discuss the architecture. What would you have built differently?

This community is growing faster than ever and I am committing to know *your* needs and wants. Why are you here? What do you want (big picture)? What is the 1 thing stopping you from building your dream *right now*?

I recently built an Instagram DM automation agent using n8n, and it ended up doing way more than just auto-replies. At first, the goal was simple — reply to incoming DMs and save time. But I kept improving it, and now it’s basically a small assistant working in the background. Here’s what it can do right now: – Automatically reply to DMs and guide the conversation– Scrape relevant websites to pull in useful data when needed– Connect with a CRM to store and organize leads– Analyze resource files (docs, PDFs, etc.) to answer questions more accurately So instead of just sending generic replies, it can actually use context and data to respond in a smarter way. I’ve been testing it on a real use case, and it definitely reduces manual work + helps not miss potential clients. I’ll drop a screenshot below so you can see it in action. If you’re running a business through Instagram and dealing with a lot of DMs, this kind of setup can save you a lot of time (and probably make you more money too). If you’re curious how this could work for your case, just send me a message — happy to walk you through it.

Looking for someone to partner with who understands WhatsApp AI chatbots very well.

Hello, Are there websites where people showcase their creations using skills, MCP, and other tools on GitHub or other marketplaces? This would give me some inspiration and, above all, help me better understand how to use them. Since there are many different search tools, I figured there was probably a reason for it. I asked some LLMs and they confirmed that Tavily can only perform searches on Google, but to enter a site and perform searches, you need to use the Brave MCP. However, there is also RPA and the Playwright MCP; here again, I'm trying to understand when to prioritize one over the other. Should I only install these two on my computer/CLI, and then the AI will choose the right one based on my prompt? But then there is another style of tool like OpenClaw or Claude Co-work; I believe OpenAI also has its own. Are these also RPA tools or are they something different? I was thinking of creating an MCP server in n8n or on my computer containing all the search tools in the world, and then, based on my prompt, it would look for the most appropriate tool. But LLMs tell me that’s not the right solution. That’s why I thought about connecting it to GitHub so it could read what the best search tools are—like a database—but that’s still not it. So, I thought about making my own database/RAG using AnythingLLM: first step, I gather all the search MCPs with Apify, and in the second step, I update this RAG with an RSS feed? This way, my CLI would have a solid knowledge base to advise me on the best tools based on what I want to do. I have different tools to resell in very different categories, and to avoid having to search for specialized sites in these categories myself, I would like to create a CLI or n8n workflow that could first find the appropriate sites and maybe even write the ads and post them on its own, and register on these sites. So, would I need to use different MCPs and perhaps even skills and other tools? Thanks.