trajeval is an open-source CLI tool for grading and evaluating agent trajectories, focusing on tool selection, argument correctness, and end state analysis. It is designed for AI researchers and developers to benchmark and improve agent performance.
trajeval sits in PulseGate's AI & ML category. It focuses on evaluating and benchmarking agent trajectories for correctness and performance in AI workflows. trajeval is an open-source project aimed at AI researchers and developers. The project is open source (MIT License). It runs on the command line.
trajeval first shipped in 2026. Among its 5 catalogued features are trajectory grading, tool selection evaluation, and argument correctness check.
Latest indexed changes and source events
Other apps tracked under the same category.
