This release adds 2 notable features for engineering teams evaluating rollout.
✓ No known CVEs patched in this version
Summary
AI summaryFixes startup HUD flash, broken microphone expansion, and improves dictation reliability.
Full changelog
Velora 0.10.14 restores fast, structured dictation and resilient recovery.
- Fixes the startup HUD flash and broken microphone expansion.
- Replaces a poisoned cleanup worker immediately after returning safe raw text.
- Makes the real startup Qwen prompt cache hit, reducing cleanup latency.
- Improves prompt-only list formatting for priorities, issues, and explicit line breaks in Default and Note modes.
- Preserves spacing across back-to-back dictations, including apps whose caret is unavailable through Accessibility.
- Invalidates stale insertion context after manual typing, clicks, Return, undo, edits, and history insertion.
- Makes tiny accidental captures fail fast instead of spending seconds in Whisper.
- Keeps audio archives atomic during cleanup recovery.
Verified with 705 Swift checks, 367 engine tests, 82.57% engine coverage, and the 14-case Qwen 3.5 4B cleanup benchmark. The DMG is Developer ID signed, notarized, stapled, and Gatekeeper accepted.
SHA-256: c795dbe7b440f4a4e4592b92ef4cb25edbe1eb6e0d83b2445702d5e062850eac
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 Velora
All releases →Beta — feedback welcome: [email protected]