A single-file Python desktop app that aggregates global RSS feeds into a retro, terminal-style GUI with automatic text-to-speech headlines. Built for fast scanning, low distraction, and hands-free updates.
- World, News, Finance, Economy, Trade
- Science, Research, Tech
- Weather (Environment Canada warnings)
- RSS + Atom support (best-effort parser)
- Top horizontal HEADLINES ribbon
- Always shows newest items
- Click to select, double-click to open
- Left pane
- Dense scan list with fixed columns
- Color-coded by category
- Right pane
- Clean detail view
- Wrapped text + clickable links
- Uses Windows System.Speech.Synthesis
- No pip installs
- Non-blocking worker thread
- Immediate stop support
- Speaks only when the newest headline changes
- No repeats
- Optional summary inclusion
- Periodic auto-refresh loop
| Key | Action |
|---|---|
1–9 |
Toggle categories |
A |
Toggle all categories |
R |
Force live refresh |
C |
Clear cache + refresh |
G |
Group by category / time |
F |
Feed status window |
Enter |
Open selected article |
S |
Speak selected item |
H |
Speak newest headline |
X |
Stop speaking |
Configurable at the top of the file: AUTO_REFRESH_SECONDS = 180 AUTO_SPEAK_NEWEST_HEADLINE_ON_CHANGE = True AUTO_SPEAK_ON_START = False AUTO_SPEAK_INCLUDE_SUMMARY = False
Behavior: Refreshes feeds automatically Tracks newest headline by ID Speaks only when it changes
Requirements Windows 10 / 11 Python 3.10+ No external Python dependencies Internet access for RSS feeds Linux/macOS: GUI works, TTS section is Windows-specific.
Run: python rss.py
Cache files will be created locally:
rssdos_cache.json rssdos_seen.json
Feed Status View Press F to open: Per-feed OK / FAIL status Item counts Active URL Error diagnostics Design Goals Fast information density Keyboard-first navigation Minimal visual noise Hands-free consumption Single-file deployability License - MIT — do what you want, attribution appreciated.
Built by Matt / VE7LTX