regress-ai is an open-source Python package that automatically converts failures encountered by LLM agents in production into a growing regression test suite. It integrates with OpenTelemetry for observability and helps developers ensure their agents become more reliable over time without manual test writing. Primarily used by AI engineers and agent developers building production LLM systems.
In the LLM eval & observability space, regress-ai takes a focused approach. Manually maintaining regression tests for LLM agent behavior in production. It is built as an open-source project for developers. regress-ai is open source under the Apache-2.0 license. It ships for the command line and API.
Arshad Vani builds and maintains regress-ai, and it first shipped in 2026. The project is developed in the open on GitHub with 81 commits in the last 90 days. Among its 4 catalogued features are Automatic Regression Suite, Production Failure Capture, and OpenTelemetry Integration. It exposes integrations via a public API.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do