winnow
GhalebDweikat/winnow
A calibrated context sieve for Claude Code: every tool result is judged by a System One model before it enters context.
Python
- 星标
- 115
- 派生
- 12
收录依据
Winnow 是 Claude Code 上下文筛选工具,默认评审模型为 Jev;README 说明配置评审密钥前不会运行,因此调用是有条件的。
GhalebDweikat/winnow
A calibrated context sieve for Claude Code: every tool result is judged by a System One model before it enters context.
Winnow 是 Claude Code 上下文筛选工具,默认评审模型为 Jev;README 说明配置评审密钥前不会运行,因此调用是有条件的。
Per-turn model & reasoning routing for Codex, driven by Jev (TypeSafe System One): picks the model, thinking depth and speed mode for every turn.
共同主题: agent
让 Agent 长出「操作杆」:用 System One 模型(Jev)承担高频判断、多模态模型当眼睛。给思路 + 可跑的参考实现(75 条离线断言,零依赖零 key)
共同主题: agent
Binary approval gate for Codex and Claude Code — every intercepted tool call is reviewed by TypeSafe JEV and composed through a versioned local policy, with scoped authorization.
共同主题: agent