This release adds 3 notable features for engineering teams evaluating rollout.
✓ No known CVEs patched in this version
Topics
+14 more
Summary
AI summaryUpdates What ships in v1.0.0, STT, and DE across a mixed release.
Full changelog
Agora Cosmica is live. This is the first public release.
Live at agoracosmica.org. Launched 2026-05-06 at 07:18 CEST (05:18 UTC).
A Living Library You Can Talk To. 30 historical figures from 2,500 years of human thought, each with a researched voice and 12 wisdom teachings. Bilingual EN + DE. Open source under AGPL-3.0. Nonprofit. No tracking.
What ships in v1.0.0
- 30 figures × 12 wisdom teachings (360 total)
- 360 narrative stories, 360 prism dialogues, 110 multi-figure council debates, all bilingual
- 6 ways to engage: Story, Wisdom, Prism, Quest, Free Talk, Council
- Live AI conversation: BYOK via OpenRouter (encrypted locally with AES-256-GCM, never transits our servers) plus Free tier via Cloudflare Worker proxy to Nebius (30 messages a day per identity, no signup)
- Self-hosted live audio on 2× Hetzner GEX130 servers in Germany: Kokoro (EN), F5 and Qwen3-TTS (DE), Faster-Whisper (STT)
- 1,660 long-form pre-recorded audio files served from Cloudflare R2 with 1-year immutable edge cache
- 90 prerendered SEO pages
- WCAG 2.2 AA, SecurityHeaders A+, SSL Labs A+, Mozilla Observatory A+ (115/100, 10/10), Hardenize clean
- GDPR, EU AI Act Article 50, German youth protection (JMStV), DDG §5 Impressum
Full release notes: CHANGELOG.md.
Built by ChipMates gemeinnützige GmbH, a German nonprofit.
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About Agoracosmica
All releases →Beta — feedback welcome: [email protected]