开源项目 Reef 发布：让智能体从反馈中持续学习改进
Reef 这个开源项目挺有意思，智能体犯过的错能自动沉淀成技能更新，不训练模型也能用，Apache 2.0 随便拿去改。
Reef 是一个开源项目，为开发者提供两种构建会学习的智能体的路径：更新模型权重，或改进模型外围的提示词、规则和技能。后者走 harness 路线，只需调用模型 API，无需本地训练 GPU。文档中给出的示例是一个编码智能体漏掉一个测试用例，反馈触发技能更新提案，发布前会与当前版本对比测试。项目以 Apache 2.0 协议开源，可自定义何为改进并追踪实际部署的版本。
I love Open Source. Its the future. Reef gives developers two ways to build agents that learn from their work: update the model's weights, or improve the prompts, rules and skills around it.
The harness route can use a model API, *without local training GPUs*.
One documented example starts with a coding agent missing a test case. Feedback triggers a proposed skill update, which is tested against the current version before publication.
That's a useful foundation for experimenting with agents that stop repeating the same mistakes. You can define what counts as an improvement and track the versions you actually deploy.
Open source under Apache 2.0. Thats the way to go.
