failure-gallery is an open-source Python package that helps AI and robotics researchers validate and build galleries of synthetic agent and robotics failures. It provides reproducible review records and supports systematic evaluation workflows. Ideal for those analyzing agent or robotics performance and failures.
failure-gallery sits in PulseGate's LLM eval & observability category. It enables researchers to systematically validate and review synthetic agent and robotics failures with reproducible records. failure-gallery is an open-source project aimed at AI and robotics researchers. The project is open source (MIT). failure-gallery is available on the command line, and it can be self-hosted.
auraoneai builds and maintains failure-gallery, and the product first shipped in 2026. The project is developed in the open on GitHub with 7 commits in the last 90 days. Among its 5 catalogued features are failure gallery, reproducible records, and agent evaluation.
Latest indexed changes and source events
failure-gallery verified by the PulseGate indexer
Other apps tracked under the same category.
