DSH Store
Trusted catalog

Publisher

作者 · patrykorwat

2 个插件。作者可回应评价,不能删差评。

screenedtools★14
open-dream-rsi#deepseek_harness

不是又一个记忆插件:ODR 是“演化世界”式(Dream-RSI)自我改进回路——每次尝试都记入发现树(Discovery Tree),任务间歇将数千个策略变体在这棵记录下来的世界上回放;策略、配方或经验只有通过反事实回放的考验才会晋升,绝不因模型自己说好就获得信任,Sentinel 还会在预算耗尽前拦截工具调用;LLM 保持冻结,可从 npm 安装 dsh-open-dream-rsi。

candidateuncategorized★0
dsh-open-dream-rsi

Let your coding agent dream: strategies must survive counterfactual replay before they earn trust. An offline self-improvement loop over an evolving world model — no fine-tuning, the LLM stays frozen. Speaks MCP. Ships its Python server inside the package