Version0.6.0
Revision11
Size1.5 MB
LicenseApache-2.0
Confinementstrict
Basecore22

Formally verified trust scoring for AI-generated text


TruthLens analyzes AI-generated text for hallucination risk using fast local
trust scoring with a formally verified core.

No API keys. No LLM calls. No cloud dependency.

Features:
- trust scoring for generated text
- claim extraction
- consistency checks
- entity extraction
- local-first CLI workflow

Built for developers, researchers, and teams evaluating LLM output quality.

Update History

0.5.0 (9)0.6.0 (11)
28 Mar 2026, 05:09 UTC
0.5.0 (9)
24 Mar 2026, 02:09 UTC

Published23 Mar 2026, 20:16 UTC

Last updated28 Mar 2026, 04:58 UTC

First seen24 Mar 2026, 02:09 UTC