在生态中，每个人都在编写产出分析报告，没有人回答“这些代理的判断到底准不准”，而是给它们一个公开的考场。

Hello——TradingAgents-CN has become the de facto standard for multi-intelligent agent research in the Chinese community, and the fork ecosystem is huge. However, the entire ecosystem output is "analytical reports", and there is always a problem that lacks the infrastructure to answer: how accurate are the judgments of these agents, settled according to the proper scoring rule?