Offload large / multi-file reads to a cheap worker model so raw files never enter Claude's context (token savings)
安装
npx skills add alinaqi/maggy --skill context-shuntnpx skills add alinaqi/maggy --skill context-shunt -a claude-code -gnpx skills add alinaqi/maggy --skill context-shunt -a codex -ggemini skills install https://github.com/alinaqi/maggy.git --path skills/context-shuntgh skill install alinaqi/maggy context-shuntgit clone --depth 1 https://github.com/alinaqi/maggy /tmp/alinaqi__maggy
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/alinaqi__maggy/skills/context-shunt ~/.claude/skills/context-shunt # Claude Code
cp -R /tmp/alinaqi__maggy/skills/context-shunt ~/.agents/skills/context-shunt # Codex, Cursor, Gemini CLI, Copilot, OpenCode