Hi everyone 
I’m working through a series of practical n8n projects aimed at people who are learning, and I share each one as a free, open repo. This is project #3: a daily finance news digest that lands in my inbox every morning.
What it does
- A Schedule Trigger starts the workflow every day
- Three RSS feeds are fetched in parallel (BBC Business, Dawn Business, CNBC Finance)
- A Merge node combines them, and a Code node keeps the newest 6 headlines per source
- Google Gemini (Basic LLM Chain) writes a short digest with the top stories, a markets section and a regional focus
- Gmail sends it to me as a clean HTML email
What I learned along the way (maybe useful if you’re building something similar)
- Pinned data on a node is great for testing, but remember to unpin it before going live, or n8n keeps replaying old results
- Limit your items before the AI node. My BBC feed alone returned 52 items, which is wasteful on free-tier limits
- Gemini sometimes wraps HTML in code fences, so I strip them in the Gmail node with a small expression
- Setting RSS nodes to “On Error: Continue” means one broken feed doesn’t kill the whole digest
Repo (workflow JSON, prompt and README with setup steps):
No keys, credential IDs or personal emails are in there. You just import the JSON, connect your own Gemini and Gmail credentials, and swap in your favourite feeds.
Heads up: it’s an AI summary of public headlines, so it’s not financial advice, and it can make mistakes. I always link back to the original articles.
I’d really value your input:
- Which sources or topics would you add (crypto, commodities, regional news, company earnings)?
- Would you rather get this by Gmail, Telegram, Slack or Notion?
- If you’ve built something similar, how do you handle duplicate stories across feeds? I’m considering a simple title-similarity check.
- What would you improve in the prompt or structure?
If you’d like to collaborate: I’m happy to team up on n8n projects, whether that’s co-building a workflow, reviewing one you’re stuck on, or turning an idea into a beginner-friendly tutorial. I come from an electrical engineering background, I’m studying computer science, and I enjoy practical automation, especially AI and finance use cases. If you have a project where an extra pair of hands would help, send me a message here or open an issue or discussion on the repo. I’ll do my best to help.
Thanks for reading, and I’m looking forward to your thoughts! 
