groundlens is an open-source CLI tool for verifiable agent triage and LLM evaluation. It uses geometric and rule-based methods to assess factual accuracy, grounding, and hallucination in AI outputs, supporting AI safety research.
In the LLM eval & observability space, groundlens takes a focused approach. It provides tools for evaluating and verifying AI agent outputs for factual accuracy and grounding. groundlens is an open-source project aimed at AI safety researchers. The project is open source (Apache-2.0). groundlens is available on the web and the command line, and it can be self-hosted.
Behind groundlens is groundlens-dev, and the product first shipped in 2026. The project is developed in the open on GitHub with 182 commits in the last 90 days. Across PulseGate's embedding index, groundlens has few near neighbours, marking it as relatively distinct. Among its 5 catalogued features are agent triage, LLM evaluation, and hallucination detection.
Latest indexed changes and source events
Other apps tracked under the same category.
