Skip to content

GhalebDweikat/winnow

110 14PythonMITUpdated 3 weeks ago
View on GitHub

A calibrated context sieve for Claude Code: every tool result is judged by a System One model before it enters context.

claude-codeclaude-code-plugincontext-managementjevllm-agentstypesafe

Our verdict: worth using

It offers a practical solution for managing context in Claude Code, which aligns directly with the engineer's work on agent tooling.

Filed under context tooling in our directory.

Similar repos