Recipes / News & Reading

Our rating: 3/5 for this setup

Daily AI digest of 100 RSS feeds, self-hosted (FreshRSS + n8n)

Each morning n8n pulls the last 24 hours of articles from a self-hosted FreshRSS, stores an embedding for each in PostgreSQL, matches them against a list of interests, and has an LLM write an email of the 10 most relevant articles, two sentences each, with links.

New Added Oct 17, 2026 · ✓ Last checked Oct 17, 2026

It broke for meBroke?
How it works: Daily AI digest of 100 RSS feeds, self-hosted (FreshRSS + n8n). Trigger: Every morning at 7 AM. The bot (FreshRSS + n8n + PostgreSQL/pgvector + Infomaniak AI): Fetches the day's articles from FreshRSS; Stores an embedding for each; Picks the 10 closest to your interests; LLM writes 2-sentence summaries. Result: Digest email around 7:30 AM. Replaces Feedly / Inoreader.HOW IT WORKSDaily AI digest of 100 RSS feeds, self-hosted (FreshRSS +…1 · TRIGGER2 · THE BOT3 · RESULTEvery morning at7 AMFreshRSSn8nPostgreSQL/pgvectorInfomaniak AIFetches the day's articles fromFreshRSSStores an embedding for eachDigest emailaround 7:30 AMREPLACESFeedly /InoreaderRATED 3/5 · HARD SETUPAppToBot
How it works: trigger: Every morning at 7 AM; the bot (FreshRSS + n8n + PostgreSQL/pgvector + Infomaniak AI): Fetches the day's articles from FreshRSS; Stores an embedding for each; Picks the 10 closest to your interests; LLM writes 2-sentence summaries; result: Digest email around 7:30 AM.HOW IT WORKS1 · TRIGGEREvery morning at 7 AM2 · THE BOTFreshRSSn8nPostgreSQL/pgvectorInfomaniak AIFetches the day's articles fromFreshRSSStores an embedding for eachPicks the 10 closest to your interestsLLM writes 2-sentence summaries3 · RESULTDigest email around 7:30 AMREPLACESFeedly /Inoreader
Diagram drawn by AppToBot from this recipe's source. It shows the flow, not a screenshot of the product.

Setup steps

  1. Run FreshRSS and group your feeds into categories
  2. Self-host n8n and PostgreSQL with pgvector
  3. Ingest: fetch the last 24h via the GReader API, dedupe, embed each article and store it
  4. Digest: embed each interest, pull the closest articles, keep the top 10
  5. An LLM writes the HTML summary; send it by SMTP

Steps summarized from the original Blog source ↗. Use “Copy” to grab them with the tools and source link, or copy an AI prompt.

How it went

The author follows about a hundred feeds (100–300 new articles a day) and now gets one email each morning with the ten most relevant, summarized. They say it's five minutes of reading.

Worth knowing

An English-only embedding model gave irrelevant matches for French; a multilingual one fixed it. n8n's built-in LangChain nodes split each article into many rows, so the author switched to plain HTTP calls and SQL.

Only what the original source ↗ says, re-read Oct 17, 2026. How we write these

Try it with AI

Paste this into ChatGPT, Claude or your own agent for step-by-step help. Prompt written by AppToBot from this recipe's data, not by the original author. It keeps drafts for your review and asks before anything is sent or paid.

Help me set up this automation using FreshRSS + n8n + PostgreSQL/pgvector + Infomaniak AI: "Daily AI digest of 100 RSS feeds, self-hosted (FreshRSS + n8n)".
Goal: replace Feedly / Inoreader.

How it should work:
- Trigger: Every morning at 7 AM
- Then: Fetches the day's articles from FreshRSS
- Then: Stores an embedding for each
- Then: Picks the 10 closest to your interests
- Then: LLM writes 2-sentence summaries
- Result: Digest email around 7:30 AM

Setup steps (summarized from the original source):
1. Run FreshRSS and group your feeds into categories
2. Self-host n8n and PostgreSQL with pgvector
3. Ingest: fetch the last 24h via the GReader API, dedupe, embed each article and store it
4. Digest: embed each interest, pull the closest articles, keep the top 10
5. An LLM writes the HTML summary; send it by SMTP

Guide me one step at a time. Ask me for anything you need (accounts, sheet or channel names) before you start.

Safety rules:
- Build and test with sample data first, and show me a sample result before anything goes live.
- Keep anything that goes to other people (emails, posts, replies, record updates) as drafts for my review until I say otherwise.
- Never send, post, buy, pay, delete or change records without my approval.
- Tell me about any cost (paid plans, API usage) before I'd incur it, and help me set a spending limit.
- Keep API keys and passwords out of shared docs and chats.
- Privacy: Article text goes to Infomaniak AI (a Swiss provider with EU infrastructure, per author); everything else stays on your server.

Based on: https://sebastienlallemand.net/en/posts/daily-digest-rss-pgvector/

Tool names link to each official site (opens in a new tab).

Cost afterNot reported (author says no paid subscription; VPS and AI API amounts not given)
SetupHard · Not reported
Time savedNot reported
SkillAdvanced

Running cost: Not reported Source mentions a modest VPS and a pay-per-use AI API but gives no amounts; author says no paid subscription. Self-hosted How we sort costs
Stack: n8n
Replace an app: AI bots that replace Feedly

App vs bot, side by side

Old app vs bot
Old app / habitBot
ToolFeedly / InoreaderFreshRSS + self-hosted n8n + pgvector similarity search + Infomaniak AI summaries by email
Monthly costNot reportedNot reported (author says no paid subscription; VPS and AI API amounts not given)
SetupNone (already in use)Not reported · Advanced
Your timeBaselineNot reported
MaintenanceHandled by vendor/you as beforeKeep the same embedding model for ingestion and search (switching means re-embedding everything); keep FreshRSS and n8n updated.
PrivacyData stays with the app vendorArticle text goes to Infomaniak AI (a Swiss provider with EU infrastructure, per author); everything else stays on your server.

Full comparison

ReplacesFeedly / Inoreader
Bot / agent stackFreshRSS + self-hosted n8n + pgvector similarity search + Infomaniak AI summaries by email
Monthly cost beforeNot reported
Monthly cost afterNot reported (author says no paid subscription; VPS and AI API amounts not given)
Setup timeNot reported
Time savedNot reported
Skill levelAdvanced
Reliability3/5: firsthand daily use with a published workflow, but SQL and embeddings to maintain yourself and no long-term data.
MaintenanceKeep the same embedding model for ingestion and search (switching means re-embedding everything); keep FreshRSS and n8n updated.
Privacy & dataArticle text goes to Infomaniak AI (a Swiss provider with EU infrastructure, per author); everything else stays on your server.
Best forSelf-hosters drowning in RSS who want one daily summary
Not forPeople who just want a reader app

Pros and cons

Pros

  • One short email instead of 100–300 new articles a day
  • Interests are a plain text list you can edit
  • Swap in Ollama or another OpenAI-compatible API

Cons

  • Several pieces to run (FreshRSS, n8n, Postgres)
  • Needs SQL and embedding know-how

Source and evidence

First-person blog post (May 2026): the author follows about a hundred feeds and gets the email every morning; the workflow is published on GitHub. No usage period or cost figures.

Read the original Blog source ↗

“Not reported” means the source doesn't say. We never fill gaps with guesses. Numbers marked “per author” or “vendor claim” are the source's own and weren't independently tested. Source last checked Oct 17, 2026.

Our rating is an editorial opinion based on the public source above, not lab testing. Spot an error? Request a correction.

Plain-text summary (for search and AI assistants)

Daily AI digest of 100 RSS feeds, self-hosted (FreshRSS + n8n). Replaces Feedly / Inoreader with FreshRSS + self-hosted n8n + pgvector similarity search + Infomaniak AI summaries by email. Cost before: Not reported. Cost after: Not reported (author says no paid subscription; VPS and AI API amounts not given). Setup: Not reported, Advanced. Time saved: Not reported. Reliability 3/5: 3/5: firsthand daily use with a published workflow, but SQL and embeddings to maintain yourself and no long-term data. Best for: Self-hosters drowning in RSS who want one daily summary. Not for: People who just want a reader app. How it went (per source): The author follows about a hundred feeds (100–300 new articles a day) and now gets one email each morning with the ten most relevant, summarized. They say it's five minutes of reading. Worth knowing: An English-only embedding model gave irrelevant matches for French; a multilingual one fixed it. n8n's built-in LangChain nodes split each article into many rows, so the author switched to plain HTTP calls and SQL. Source: https://sebastienlallemand.net/en/posts/daily-digest-rss-pgvector/

Related recipes

More: all News & Reading recipes · Best AI agents to replace Zapier · AI agents that replace browsing and search · n8n vs Make: AI agent recipes compared · AI bots that replace Feedly

Compare with another recipe ← All recipes

Questions, corrections or ideas?

Spotted an error, want a recipe covered, or have a partnership idea? Send us a message and we'll get back to you.

Contact us

Starting your own newsletter? Try beehiiv ↗

What is beehiiv?

beehiiv is a newsletter platform: you write and send emails, host a newsletter website, and grow and monetize a list from one place. We've used it for our own newsletter.

beehiiv says (vendor claims, not tested by us):

  • Free plan with up to 2,500 subscribers and unlimited sends
  • Built-in website, recommendation network and analytics
  • 0% cut of paid subscription revenue (Stripe fees apply)

Good for: starting or growing an email newsletter. Not ideal for: email automations or removing beehiiv branding on a $0 budget (those need Lite or Pro).

Free plan: $0, no card needed. Paid plans from $49/mo billed annually (Lite, up to 1,000 subscribers). People who sign up through our link get a 14-day trial plus 20% off their first 3 months, per beehiiv's partner page. Facts as of Oct 2026, from beehiiv's own website. More about beehiiv →

Press / anywhere to search. Open full search

Send us a bot that replaced an app

Three quick fields. We check every link by hand and only publish what the source supports.

Please don't enter an email address here; we never publish emails. Stored for up to 12 months. See Privacy and Terms.