ecosystem-primer

Guides framework selection among LangChain, LangGraph, and Deep Agents before writing agent code.

1.2k|90|Updated Jan 22, 2026
One-click install
npx skills add https://github.com/langchain-ai/langchain-skills --skill ecosystem-primer
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ecosystem-primer
Source: https://github.com/langchain-ai/langchain-skills/tree/main/config/skills/ecosystem-primer
Command: npx skills add https://github.com/langchain-ai/langchain-skills --skill ecosystem-primer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Choosing between LangChain, LangGraph, and Deep Agents is confusing because the tools overlap and their APIs change frequently. This Skill acts as the mandatory first step for any LangChain-ecosystem agent project, ensuring you pick the right layer, configure the environment correctly, and load the right follow-up skill before writing code.

Core Features & Use Cases

  • Framework Decision Logic: An ordered decision table routes tasks to Deep Agents (planning, memory, subagents), LangGraph (custom control flow), or LangChain (single-purpose agents and chains).
  • Environment Setup: Provides the current LangSmith environment variables (LANGSMITH_API_KEY, LANGSMITH_TRACING, LANGSMITH_PROJECT) for observability from day one.
  • Docs Navigation: Explains how to query docs.langchain.com via the Docs MCP server or the llms.txt index for up-to-date API references.
  • Skill Routing: Directs you to the correct layer-specific skill (e.g., langchain-fundamentals, langgraph-persistence, deep-agents-core) or quickstart for your project.
  • Use Case: You are starting a new project that needs a long-running agent with file management and subagent delegation. This Skill identifies Deep Agents as the right layer, sets up tracing, and points you to deep-agents-core before any code is written.

Quick Start

Ask your agent to help you choose between LangChain, LangGraph, and Deep Agents for your new agent project and set up the environment.

Frequently Asked Questions about ecosystem-primer

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I choose between LangChain, LangGraph, and Deep Agents?

Evaluate needs in order: use Deep Agents if the task requires planning, file management, persistent memory, or subagent delegation; LangGraph for custom control flow like deterministic loops and branching; LangChain for single-purpose agents with a fixed tool set or simple chains.

What is the difference between LangChain and LangGraph?

LangChain is a framework providing model, tool, and agent-loop abstractions via create_agent. LangGraph is a lower-level runtime for durable execution and custom control flow using StateGraph with explicit nodes and edges. LangChain agents run on top of LangGraph.

Can I combine Deep Agents with LangGraph in one project?

Yes, the layers compose. A compiled LangGraph graph can be registered as a named subagent inside Deep Agents, and the orchestrator delegates to it via the task tool. LangChain tools and retrievers also work inside LangGraph nodes and Deep Agents tools.

What environment variables does LangSmith tracing require?

Set LANGSMITH_API_KEY, LANGSMITH_TRACING=true, and LANGSMITH_PROJECT. These are the current variable names; older names no longer work. Model provider keys like ANTHROPIC_API_KEY or OPENAI_API_KEY are set separately as needed.

When should I not use Deep Agents?

Avoid Deep Agents when the task is simple enough for a single-purpose LangChain agent, or when you need precise hand-crafted control over every graph edge, in which case use LangGraph directly.