AgentKit Benchmark Report

Timestamp: 2026-06-17T03:26:25.072956+00:00

Version: 0.1.0

Overall Score: 98.0%

Summary: 50/51 tests passed (1 failed) across 7 dimensions.

Dimension Results

Dimension	Total	Pass	Fail	Score
preprocessing	15	14	1	93.3%
overfitting	3	3	0	100.0%
efficiency	5	5	0	100.0%
tool_search	10	10	0	100.0%
event_model	6	6	0	100.0%
spec_management	7	7	0	100.0%
verification	5	5	0	100.0%
OVERALL	51	50	1	98.0%

[preprocessing] skill_prefix_direct

expected: skill_react

actual: direct_chat