In this video, I show you how I built a no-code Photoshop AI agent inside n8n that can combine and edit images, search through all your files for raw content or generated images, and even rename them automatically. Powered by Google’s new state-of-the-art Nano Banana image model, this agent is a game-changer for ad creatives and UGC content, making high-quality image editing and management faster and easier than ever. As always, you can get all the free resources you need to copy this exact system in this post :)

Assuming you already know what MCP is. More and more tools, IDEs, CLIs, even terminals like Warp, now speak MCP. Which is cool… until you realize you’re juggling 20+ MCP servers, each with its own OAuth setup and API keys. Enter Rube (rube.app). Recently (re)launched on Product Hunt(link), it’s basically a “super MCP” that claims to connect you to 500+ apps through one server. Just drop it into your tool and let it handle the messy app integrations. But here’s the problem: docs are thin, and I couldn’t find a real tutorial on using Rube together with n8n. So I made one. For visual learners, here’s the video and a text version at https://www.linkedin.com/pulse/supercharging-n8n-rube-one-mcp-all-jove-zhong-0yugc/

Hey everyone 👋 I’ve been thinking about experimenting with a Google Chrome Extension that connects to n8n as the backend. For example: - Having a small Chrome extension button to quickly trigger an n8n workflow. - A use case might be something like posting or commenting on LinkedIn (where the extension collects input → sends it to n8n → workflow executes the action, e.g. posting a comment, saving data, or calling an API). I’m curious if anyone here has: - Tried integrating a browser extension with n8n before? - Used the extension as a simple frontend to trigger workflows? - Faced any authentication/security challenges connecting the two? Would love to hear if anyone experimented with this or if there are best practices/examples out there 🙌

A relatively new tool: https://rube.app/ and PH page: https://www.producthunt.com/products/rube "Rube is a universal MCP that enables you to take actions across 600+ applications from within your AI chat, be it Cursor, Claude, VS Code, etc. It manages authentication and tool selection seamlessly and securely within the chat." You can configure this in n8n, in LLM chat clients, or coding agents like Claude Code. I prefer self-hosting n8n with my private app for Google. But if you can trust rube, this seems to be all-in-one cloud solutions to automate many things. I guess if you enable one rube integration, such as Slack, then a few new MCP tools will be available via the http transport

Just got level2 and now I can post and share my workflows 🎉 Last month I wrapped my time at a startup and started looking for opportunities. The best way is internal-referral for sure, but checking open positions in LinkedIn or company websites can help you apply more positions and get more onsite/remote interviews to strength your interview muscle 💪 The goal is to get email/slack/text notification when a certain company is hiring for a certain role or at certain location. This doesn't need to be real-time. Daily check is good enough, since there won't be so many new gigs posted every day😂 I will share my sample workflow in the end, but you almost have to customize it so please read on. Let's start from a simple case, you want to find out whether there is any new position for Amazon or AWS in Vancouver (yes, I live here 🇨🇦). You go to amazon.com and find their career site https://amazon.jobs/ Click the link for "Software Development" and landed https://amazon.jobs/content/en/job-categories/software-development I would suggest before working on the automation, do this manually a few times to make sure this worths the automation effort. Choose a few filters, such as Canada, BC, Vancouver, Full time, then you got 38. I am looking for manager roles, so I typed "Manager" in search bar and got a bit less. Sometime it's a single page, sometimes there can be multiple pages for the positions. But you don't have to automate the pagination or parse the DOM. If you have a little bit JavaScript background, you can try to figure out whether there is data feed behind the page. So open the JavaScript Console, go to Nextwork tab, filter by Fetch/XHR, reload the page, and guess which request contains the job list. Usually with 10KB or more response size. Then check the request details, is that GET or POST, what's the URL, what's the payload, and also whether by default the page size is 10, you probably can change it to 100 to get first 100 times without pagination.