langgraph-docs
๐ฏSkillfrom langchain-ai/deepagents
|Batteries-included agent harness by LangChain providing a ready-to-run agent with built-in planning, filesystem access, shell execution, sub-agent delegation, and automatic context management.
Overview
LangGraph Docs is a skill from the Deep Agents repository by LangChain AI, a batteries-included agent harness that provides a ready-to-run agent with planning, filesystem access, shell execution, sub-agents, and context management out of the box. Built on LangGraph, it returns compiled graphs compatible with streaming, Studio, checkpointers, and the full LangGraph feature set.
Key Features
- Batteries-Included Agent - Planning (
write_todos/read_todos), filesystem operations, shell access with sandboxing, and sub-agent delegation all included by default - Context Management - Auto-summarization when conversations get long and automatic saving of large outputs to files for efficient context window usage
- Provider Agnostic - Works with Claude, OpenAI, Google, or any LangChain-compatible model through
init_chat_model - CLI with Advanced Features - Terminal interface with conversation resume, web search, remote sandboxes (Modal, Runloop, Daytona), persistent memory, and human-in-the-loop approval
- MCP Support - Integration with Model Context Protocol through
langchain-mcp-adaptersfor connecting to external tool servers
Who is this for?
This skill is designed for developers building AI agents with LangGraph who need documentation and patterns for the Deep Agents framework. It is particularly useful for teams that want a production-ready agent harness with built-in planning, file management, and context optimization rather than building these capabilities from scratch.
Same repository
langchain-ai/deepagents(18 items)
Installation
npx vibeindex add langchain-ai/deepagents --skill langgraph-docsnpx skills add langchain-ai/deepagents --skill langgraph-docs~/.claude/skills/langgraph-docs/SKILL.mdSKILL.md
More from this repository10
Performs web research tasks within the Deep Agents framework, an opinionated agent harness that provides ready-to-run agents with built-in prompts, tools, and context management.
Batteries-included agent harness by LangChain providing a ready-to-run agent with built-in planning, filesystem access, shell execution, sub-agent delegation, and automatic context management.
Batteries-included agent harness by LangChain providing a ready-to-run agent with built-in planning, filesystem access, shell execution, sub-agent delegation, and automatic context management.
Batteries-included agent harness by LangChain providing a ready-to-run agent with built-in planning, filesystem access, shell execution, sub-agent delegation, and automatic context management.
Batteries-included agent harness by LangChain providing a ready-to-run agent with built-in planning, filesystem access, shell execution, sub-agent delegation, and automatic context management.
Batteries-included agent harness by LangChain providing a ready-to-run agent with built-in planning, filesystem access, shell execution, sub-agent delegation, and automatic context management.
Explores and maps database schemas and API structures for rapid comprehension, from Deep Agents, LangChain's batteries-included agent harness.
A Deep Agents skill for data visualization tasks, part of LangChain's open-source agent harness that supports sub-agents, filesystem access, shell commands, and persistent memory across any LLM with tool calling.
Part of the Deep Agents library by LangChain, a batteries-included agent harness that provides planning, filesystem access, shell execution, sub-agent delegation, and auto-summarization out of the box with smart defaults.
Deep Agents is an opinionated, batteries-included agent harness by LangChain that provides planning, filesystem access, shell commands, sub-agents, and context management out of the box. Built on LangGraph, it supports any LLM with tool calling and can be customized with your own tools, prompts, and models.