skills.rest
langchain-ai
langchain-aiOFFICIAL
Explore all skills by this author →
Website𝕏@LangChainGitHub

langgraph-python-quickstart

Scaffolds a minimal local LangGraph agent in Python following the official quickstart.

1.2k|90|Updated Jan 22, 2026
Search Tags:#Python#langgraph#Langchain#Quickstart#Agent#Scaffold#chat model
One-click install
npx skills add https://github.com/langchain-ai/langchain-skills --skill langgraph-python-quickstart
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: langgraph-python-quickstart
Source: https://github.com/langchain-ai/langchain-skills/tree/main/config/skills/langgraph-python-quickstart
Command: npx skills add https://github.com/langchain-ai/langchain-skills --skill langgraph-python-quickstart

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Setting up a first LangGraph agent often involves guessing at APIs, hardcoded model choices, and cluttered project directories. This Skill walks through the official LangGraph Python quickstart so you get a working local agent with the correct current API and your preferred model provider.

Core Features & Use Cases

  • Official quickstart alignment: Fetches the live LangGraph quickstart docs and implements the calculator/math agent using the Graph API rather than relying on memorized APIs.
  • Model-agnostic setup: Prompts for a provider:model string (e.g. openai, anthropic, google_genai) and wires it via init_chat_model, with provider-specific constraints handled.
  • Clean local environment: Creates a dedicated directory, keeps the API key in a gitignored .env file, and installs only the required packages.
  • Use Case: You want to try LangGraph locally for the first time — the Skill scaffolds a new langgraph-agent directory, configures your chosen model, runs the example "Add 3 and 4.", and shows the output.

Quick Start

Ask your agent to scaffold a minimal local LangGraph agent in Python using the official quickstart with your preferred model provider.

Frequently Asked Questions about langgraph-python-quickstart

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

FAQPage Schema
How do I build a LangGraph agent in Python locally?▼

Follow the official LangGraph Python quickstart, which builds a calculator/math agent using the Graph API. Create a new directory, install the quickstart packages plus your model provider package, set the API key in a .env file, and run the example.

Which models can I use with a LangGraph agent?▼

LangGraph works with any LangChain chat model via init_chat_model using a provider:model string, such as openai:gpt-5.5, anthropic:claude-sonnet-5, or google_genai:gemini-2.5-flash-lite. For Claude Sonnet 5+, omit temperature, top_p, and top_k parameters.

Should I use the Graph API or Functional API in LangGraph?▼

The quickstart prefers the Graph API path over the Functional API unless you specifically ask otherwise. For a higher-level agent interface, use LangChain's create_agent instead of building the graph directly.

Do I need LangSmith or Tavily keys for the LangGraph quickstart?▼

No. The only secret required is your model provider's API key stored in a gitignored .env file. LangSmith and Tavily are optional and only added if you explicitly request them.

What are the next steps after the LangGraph quickstart?▼

After running the example successfully, move on to the langgraph-fundamentals material for deeper concepts. If you want a higher-level abstraction, switch to LangChain's create_agent API.

Related Skills

Explore More Software Engineering Skills→
Varnan-TechVarnan-Tech
581

geo-gap-fixer

Audit brand visibility on LLMs and generate a GEO action plan.

Official
Advanced
amiteamite

langgraph

Orchestrate durable, stateful agent workflows with LangGraph nodes and edges.

Community
Advanced
cuba6112cuba6112

langchain-agents

Develop LLM agents with LangChain and LangGraph for tool integration and state management.

Community
Intermediate
jackjin1997jackjin1997
11

LangChain Fundamentals

Create LangChain agents with create_agent, @tool, and middleware.

Community
Intermediate