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

by GhalebDweikat · Claude Skill · ★ 115

Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h

How to install winnow

git clone https://github.com/GhalebDweikat/winnow.git
claude plugin install winnow@winnow

Taken from the project's README on GitHub. Check the security audit before running it.

Security audit verdict: SAFE · quality 73/100

None of the 42 rules matched — no credential harvesting, no data exfiltration, no curl-pipe-shell installer. Scanned against 42 published rules built on the SlowMist agent-security taxonomy, refreshed every 8 hours. Full audit →

The quality score combines nine heuristic checks of the repo and its README: completeness, clarity, specificity, examples, structure, agent readiness, step-by-step content, instruction quality and security notes. It rates how well the project explains itself to an agent, not the code.

claude-codeclaude-code-plugincontext-managementjevllm-agentstypesafe

Quick Facts

Stars115
Forks14
LanguagePython
CategoryClaude Skill
LicenseMIT
Quality Score73/100
Open Issues2
Last Updated2026-10-06
Created2026-09-16
Platformsclaude-code, python
Est. Tokens~19k

About winnow

winnow A calibrated context sieve for Claude Code. Every large , , or result is judged before it enters Claude's context. Blocks the judge is confident you don't need are replaced with a three-line stub: what was hidden, a one-paragraph summary from a cheap model, and a key that restores the full text on demand. Nothing is lost; it just stops costing tokens until you ask for it. Terms used below.

Compatible Skills

These tools work well together with winnow for enhanced workflows:

  • jevmeter — semantic(0.29)+complementary+rare_topics+same_lang+similar_pop+shared_platform (69%)
  • jev-codex-router — semantic(0.28)+complementary+rare_topics+same_lang+similar_pop+shared_platform (69%)
  • HA-Jev — semantic(0.22)+complementary+rare_topics+same_lang+similar_pop+shared_platform (67%)
  • jev-drone — semantic(0.28)+complementary+rare_topics+same_lang+similar_pop+shared_platform (64%)

winnow alternative? Top 6 similar tools

Looking for a winnow alternative? If you're comparing winnow with other claude skill tools, these 6 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.

  • Canny by qkal · ⭐ 50

    Stops AI coding agents from claiming work is done without evidence. Deterministic hooks decide, TypeSafe's Jev

  • claude-elixir-phoenix by oliver-kriska · ⭐ 569

    Claude Code plugin for Elixir/Phoenix/LiveView — 26 specialist agents, Iron Laws enforcement, and Tidewave MCP

  • adversarial-spec by zscole · ⭐ 557

    A Claude Code plugin that iteratively refines product specifications by debating between multiple LLMs until a

  • gmickel-claude-marketplace by gmickel · ⭐ 547

    Claude Code plugins for reliable AI coding. Flow-Next: plan-first workflows, Ralph autonomous mode (overnight

  • skills by posit-dev · ⭐ 543

    A collection of Claude Skills from Posit

  • claude-code-hooks by karanb192 · ⭐ 543

    🪝 Claude Code hooks + an installable plugin marketplace: safety, cost, observability, productivity.

More Claude Skill Tools

Explore other popular claude skill tools:

View all Claude Skill tools →

Popular Python Agent Tools

How this security grade is produced

Grades come from a rule-based scan built on the SlowMist agent-security taxonomy: 42 published rules, covering credential harvesting, data exfiltration, curl | sh installers and more. It is a first-layer scan, not a manual audit — we say so rather than overstate it.

The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.

Sources & who's responsible:

View on GitHub → Browse Claude Skill tools