INVOKE THIS SKILL when you need human-in-the-loop approval, custom middleware, or structured output. Covers HumanInTheLoopMiddleware for human approval of dangerous tool calls, creating custom middleware with hooks, Command resume patterns, and structured output with Pydantic/Zod.
安装
npx skills add langchain-ai/langchain-skills --skill langchain-middlewarenpx skills add langchain-ai/langchain-skills --skill langchain-middleware -a claude-code -gnpx skills add langchain-ai/langchain-skills --skill langchain-middleware -a codex -ggemini skills install https://github.com/langchain-ai/langchain-skills.git --path config/skills/langchain-middlewaregh skill install langchain-ai/langchain-skills langchain-middlewaregit clone --depth 1 https://github.com/langchain-ai/langchain-skills /tmp/langchain-ai__langchain-skills
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/langchain-ai__langchain-skills/config/skills/langchain-middleware ~/.claude/skills/langchain-middleware # Claude Code
cp -R /tmp/langchain-ai__langchain-skills/config/skills/langchain-middleware ~/.agents/skills/langchain-middleware # Codex, Cursor, Gemini CLI, Copilot, OpenCode