Just curious what people are using n8n for.
I just finished setting up a workflow that sends me a Telegram message every night about photography opportunities for the next day. It puts together weather data, POIs (which I defined for my location), sun/moon position, milky way visibility, cloud cover, etc. The message then simply tells me if it’s worth it going out in the morning.
You must log in or # to comment.
A governmental-ish site I’m required to use doesn’t push notifications as mails, so you have to login daily to check for updates. Updates may happen multiple times daily or once a month. I automated my server to access the site once a day with my credentials, screenshot the notifications, parse them with ocr, and send myself a mail.
What do you use for OCR parsing?
The data is non critical and doesn’t contain indentifying info so I use ocr.space API. You could probably find ways to use the tesseract libraries locally.
Why screenshot and parse? Can’t you just parse the html directly?
Since the dawn of LLMs it’s virtually impossible to scrape web content. Headless browsers have become basically useless. I actually have to automate keyboard inputs to simulate the navigation. I could maybe try to write the javascript cache to file but honestly it’s just faster that way.
What why, I’m scraping html just fine
Acronyms, initialisms, abbreviations, contractions, and other phrases which expand to something larger, that I’ve seen in this thread:
Fewer Letters More Letters DNS Domain Name Service/System HTTP Hypertext Transfer Protocol, the Web IP Internet Protocol NAT Network Address Translation nginx Popular HTTP server
4 acronyms in this thread; the most compressed thread commented on today has 14 acronyms.
[Thread #242 for this comm, first seen 17th Apr 2026, 11:20] [FAQ] [Full list] [Contact] [Source code]
I’m not willing to run something that risky on my own infrastructure, therefore I wont run it. I can do all of those things with a handful of scripts or Home Assistant anyway.
May I ask why you consider n8n risky?
Just look at the recent vulnerabilities it had. It just had a CVE 10 in January and March of this year.
Almost all of them affect authenticated users. The only ones that allow unauthenticated users access are for old versions. So to be honest: I am not that much concerned.
You aren’t concerned about a severity 10 vuln? Those are found in software with exceptionally bad qa and abandonware. That should give you pause about how hard n8n is “going fast and breaking things”.
Well, this “authenticated user” could be your fucking LLM and this is really concerning. Besides that a single CVE 10 is also very concerning no matter how old the affected version is.
I’m not a slop jockey and I’m watching people plug these kinds of things into public data and having it do operations with financial/business impacts.
I’d rather employ a gremlin to write scripts.
I’m not a slop jockey
Are you aware that you are not required to run AI with n8n? There are two flavors: With AI and without.
lowcode environments are the O.G slop.
Is there a version of n8n without AI generated code?
If you are asking if you can run n8n without AI, yes. That’s what I run. If you are asking whether n8n is free of any AI generated code used in it’s creation, I couldn’t speak to that.
I see. It uses AI generated code, I just checked.
This is going to be a very difficult hill to die on as time goes on…
Tons of things 😊
- Built quite a few personal agents, long before OpenClaw or whatever that is
- Automated Social Media Images
- Set up a bot on Manifold Markets
- Created entire processes in our company using it for Sales, Marketing and Customer Support
- Automatically update the CRM using Macrodroid and n8n
- tons of other playing around.
One of the most amazing pieces of software I’ve ever used.
Can you elaborate on the agents? How do you handle persistant memory, what triggers them? What are they for? ☺️
n8n offers some cool agent features out of the box.
Like, if I talk about my personal agent is connected to Telegram, it can access my emails, my calendar, it can do web searches, deep research, it can access a NocoDB instance. etc. It also records a journal, any food diaries, it can take handwritten notes and convert them to text and store in my obsidian vault.
For memory, I use the standard one from n8n but for long term memory, I use NocoDB or some vector storage (I forget the name).
Honestly, it’s very easy to get started and play around with, so I recommend just setting up an agent and just having a go.
