TagesByte/README.md

Tagesbyte

Your daily tech news in bytes. A Flutter app that collects the day's tech news from German and English RSS/Atom feeds and uses AI to summarize single articles or write a daily briefing. It's meant to replace the n8n → Discord workflow.

Features (prototype)

• Today: all articles from the enabled sources in the last 12/24/48 h, newest first. Filter by language or source, pull to refresh.
• Learn: articles from the enabled Learning sources (engineering blogs, system design explainers, AI research write-ups) from the last 14 days, newest first. Learning sources aren't part of the Briefing.
• Reading in the app: tapping an article opens a clean reader view (text, headings, lists, quotes, images) without ads or cookie banners. The globe icon switches to the original page in an in-app WebView, for paywalls or sites that block reader view. Summarize with AI uses the same downloaded text, so the summary is based on the full article, not just the RSS teaser. Links in the briefing open the same way.
• Briefing: AI picks the 6–10 most important stories of the day, merges coverage from different outlets, and links every topic to its sources. Every briefing is saved on the device (briefings.json in the app's private storage); the history button lists them by month, tap to read, swipe to delete (with undo).
• Paid articles and ads hidden: heise+ and Golem (g+) are recognized by their title; SPIEGEL+ and t3n Pro by a marker in the article page, which the app checks once per new article in the background (it reads only until the marker, and remembers results for 14 days). Sponsored posts (Anzeige:) are hidden too. The feed header shows how many were hidden; both filters can be turned off.
• Ad and cookie-banner blocking in the original-page view: requests to ~80 ad/tracker hosts and the common consent managers (Sourcepoint, OneTrust, consentmanager, Usercentrics, Didomi, Cookiebot, …) are blocked, and a stylesheet hides leftover ad slots and banners. Nothing is accepted, so sites treat it like a rejection. Golem redirects to a separate consent page instead of showing a banner; with "Accept cookie walls" on (default), the app makes the same request as Golem's own "Zustimmen" button (/abo/setconsentcookie.php, valid for a day) and reuses the cookie for reader view and the in-app page. With it off, the app shows the teaser and an "Open website anyway" option.
• Sources: 23 curated feeds (13 German, 10 English), toggle each one, or add any RSS/Atom feed (it's validated before being added).
• Dark mode: System / Light / Dark in Settings (applies instantly). The original-page view follows it too: when the phone is dark, sites use their own dark design or Android darkens them; when only the app is dark, the app inverts the page colours and keeps photos and videos as they are.
• Settings: AI provider (off / Claude API / n8n webhook), model, summary language (German/English, whatever the source language), time window, paid/ad filters, ad and cookie-banner blocking.

Running

flutter pub get
flutter run            # pick an Android emulator/device
flutter test           # unit tests (parser, Claude request/response handling)
LIVE_FEEDS=1 flutter test test/live_feeds_test.dart   # checks every real feed

Platform notes:

• Android / iOS: work as-is (Android 6.0+).
• Windows desktop: requires Visual Studio with the "Desktop development with C++" workload. Reader view works; "original page" opens the browser because the WebView plugin has no Windows support.
• Web: browsers block cross-origin feed requests (CORS). Set a CORS proxy prefix in Settings (only shown on web), or use the app on mobile/desktop.

AI setup

Option A: Claude API (direct)

1. Create a key at https://console.anthropic.com/settings/keys.
2. Settings → AI summaries → Claude API → paste the key → Save.

The key is stored in the platform keystore (flutter_secure_storage) and only sent to api.anthropic.com. Model choices:

Model	Use it for
claude-opus-5-5 (default)	Best briefing quality
claude-sonnet-5-5	Balanced cost and quality
claude-haiku-5-5	Cheapest; good enough for single-article summaries

Requests use low effort (summarizing is routine work) and, for the briefing, structured JSON output so the result maps reliably to topics and links. On Opus and Sonnet the app also enables Anthropic's server-side refusal fallback (fallbacks: "default"): if a safety classifier declines a request, the API retries it on a fallback model.

Shipping the app to other people with your key inside isn't safe. For personal use a key you entered yourself is fine; for distribution, route requests through your own backend (e.g. the n8n option below).

Option B: Your n8n workflow

Settings → AI summaries → n8n webhook → enter the production URL of a Webhook trigger node (method POST, "Respond: Using 'Respond to Webhook' node"). The app sends JSON and expects JSON back.

Daily briefing

// Request
{
  "action": "daily_briefing",
  "language": "de",                       // "de" | "en"
  "articles": [
    {
      "id": "a1",
      "title": "…",
      "url": "https://…",
      "source": "heise online",
      "language": "de",
      "published": "2026-10-11T07:30:00.000Z",
      "excerpt": "…"
    }
  ]
}

// Response, option 1 (structured; topics link back to the articles)
{
  "intro": "One or two sentences on the day.",
  "topics": [
    { "title": "…", "summary": "…", "article_ids": ["a1", "a7"] }
  ]
}

// Response, option 2 (plain text, e.g. your existing Discord message)
{ "text": "…" }

Single article

// Request
{
  "action": "summarize_article",
  "language": "de",
  "article": { "title": "…", "url": "…", "source": "…", "published": "…", "text": "extracted article body" }
}

// Response
{ "summary": "…" }       // a plain-text body also works

Use a Switch node on {{$json.body.action}} to route the two cases to your existing AI nodes.

News sources

Every feed URL below was checked to load and parse. The ★ sources are enabled by default: a small set of high-signal News sources. The others stay in the catalog to switch on in the Sources tab.

	Source	Feed	Focus
🇩🇪 ★	heise online	heise.de/rss/heise-atom.xml	Broad IT news, high volume (~150/day)
🇩🇪 ★	Golem.de	rss.golem.de/rss.php?feed=ATOM1.0	IT news for professionals
🇩🇪 ★	The Decoder	the-decoder.de/feed/	AI news: models, research, industry
🇩🇪	t3n	t3n.de/rss.xml	Digital business, startups, AI
🇩🇪	ComputerBase	computerbase.de/rss/news.xml	Hardware, software, gaming
🇩🇪	SPIEGEL Netzwelt	spiegel.de/netzwelt/index.rss	Mainstream news magazine view
🇩🇪	ZEIT Digital	newsfeed.zeit.de/digital/index	Digital policy and society
🇩🇪	netzpolitik.org	netzpolitik.org/feed/	Digital rights, privacy
🇩🇪	heise Security	heise.de/security/rss/news-atom.xml	Security section of heise online (turn heise online off to use it)
🇩🇪	heise Developer	heise.de/developer/rss/news-atom.xml	Developer section of heise online (turn heise online off to use it)
🇩🇪	Caschys Blog	stadt-bremerhaven.de/feed/	Consumer tech, smartphones
🇩🇪	derStandard Web	derstandard.at/rss/web	Austrian daily's tech section
🇩🇪	Borns IT- und Windows-Blog	borncity.com/blog/feed/	Windows, patches, admin topics
🇬🇧	The Verge	theverge.com/rss/index.xml	Consumer tech, policy
🇬🇧 ★	Ars Technica	feeds.arstechnica.com/arstechnica/index	In-depth tech and science
🇬🇧	TechCrunch	techcrunch.com/feed/	Startups, funding
🇬🇧 ★	Hacker News	hnrss.org/frontpage?points=150	Front page stories with 150+ points
🇬🇧 ★	Techmeme	techmeme.com/feed.xml	The day's top tech stories, picked by editors
🇬🇧 ★	Lobsters	lobste.rs/rss	Computing links, voted by a programmer community
🇬🇧 ★	Simon Willison	simonwillison.net/atom/entries/	LLMs and AI tools, tried out hands-on
🇬🇧	MIT Technology Review	technologyreview.com/feed/	AI, biotech, climate
🇬🇧	The Register	theregister.com/headlines.atom	Enterprise IT
🇬🇧	BleepingComputer	bleepingcomputer.com/feed/	Security, malware

Reader view works for the publishers' own articles except ZEIT Digital, which blocks non-browser requests, so the app shows the teaser and offers the original page. Hacker News and Lobsters link to articles all over the web, so reader view depends on the linked site. Techmeme links to its own summary page, which only has the teaser; the original page links on to the full story.

Learning sources

Evergreen reading for building skills: research, engineering depth and system design rather than product news. The Learn tab shows their posts from the last 14 days. All are English; ★ marks the ones enabled by default.

	Source	Feed	Focus
🇬🇧 ★	ByteByteGo	blog.bytebytego.com/feed	System design explained with diagrams
🇬🇧 ★	System Design One	newsletter.systemdesign.one/feed	System design case studies of real systems
🇬🇧 ★	The Pragmatic Engineer (blog)	blog.pragmaticengineer.com/rss/	How big tech and startups build software (free posts)
🇬🇧 ★	Martin Fowler	martinfowler.com/feed.atom	Software architecture, refactoring, delivery
🇬🇧 ★	Julia Evans	jvns.ca/atom.xml	Systems, networking and Unix tools, explained plainly
🇬🇧 ★	Dan Luu	danluu.com/atom.xml	Long, data-driven essays on computers and engineering
🇬🇧 ★	The Cloudflare Blog	blog.cloudflare.com/rss/	Networking, security, incident write-ups
🇬🇧 ★	Netflix Technology Blog	netflixtechblog.com/feed	Distributed systems, data, streaming at scale
🇬🇧 ★	Engineering at Meta	engineering.fb.com/feed/	Infrastructure, AI, production engineering
🇬🇧 ★	GitHub Blog: Engineering	github.blog/engineering/feed/	How GitHub builds and runs GitHub
🇬🇧 ★	Latent Space	www.latent.space/feed	AI engineering: interviews and deep dives
🇬🇧 ★	Import AI	importai.substack.com/feed	Weekly AI research digest by Jack Clark
🇬🇧 ★	Interconnects	www.interconnects.ai/feed	Language model training and research
🇬🇧 ★	Ahead of AI	magazine.sebastianraschka.com/feed	LLM research explained by Sebastian Raschka
🇬🇧 ★	Lilian Weng	lilianweng.github.io/index.xml	Thorough surveys of machine learning topics
🇬🇧	InfoQ	feed.infoq.com/	Architecture and engineering talks, high volume

Left out on purpose: The Pragmatic Engineer newsletter (mostly paid), One Useful Thing (applied AI), vendor blogs (OpenAI, DeepMind, Hugging Face) and Chip Huyen (inactive since January 2025). Some feeds are long (Dan Luu ~130 posts, Lilian Weng ~50); the 14-day window keeps the list short.

Project layout

lib/
  main.dart                    App entry, theme
  data/default_sources.dart    Curated feed list
  models/                      Article, NewsSource, DailyBriefing, AppSettings
  services/
    feed_parser.dart           RSS 2.0 / RSS 1.0 / Atom parser, date parsing
    feed_service.dart          Parallel fetching, dedupe, time window
    article_extractor.dart     Article page → reader blocks + text for the AI
    summarizer.dart            Summarizer interface and shared prompt input
    claude_summarizer.dart     Claude Messages API over HTTP
    n8n_summarizer.dart        n8n webhook bridge
    storage.dart               Settings, sources, briefing persistence
    paywall_service.dart       Paid/sponsored detection (titles + page marker)
    content_blocking.dart      Ad/tracker/consent blocklists for the WebView
    consent_service.dart       Accepts consent walls (Golem) and keeps the cookies
  state/app_state.dart         ChangeNotifier holding app state
  ui/                          Screens (Today, Article, Briefing, Sources, Settings)