detective-ci provides deterministic blame-level golden replay capabilities and acts as a CI gate for the Agent Detective observability suite. It enables reproducible testing of multi-agent runs by replaying traces and enforcing quality thresholds in continuous integration pipelines. Targeted at developers maintaining complex LLM agent systems.
In the LLM eval & observability space, detective-ci takes a focused approach. It focuses on automating quality regression detection and CI enforcement for multi-agent AI systems. detective-ci is an open-source project aimed at developers. The project is open source (BUSL-1.1). detective-ci is available on the command line.
detective-ci first shipped in 2026. Among its 3 catalogued features are Golden Replay, CI Gate, and Blame Analysis. It exposes integrations via a public API.
Latest indexed changes and source events
Other apps tracked under the same category.
