langgraph-typescript-quickstart

langchain-ai/langchain-skills

Scaffold a minimal LangGraph agent in TypeScript from the official quickstart.

What is langgraph-typescript-quickstart?

Quickly set up a local LangGraph agent in TypeScript by following the official quickstart guide. Use this when you want to build or test a LangGraph agent with a calculator/math example, choosing your preferred LLM provider.

Scaffolds a minimal LangGraph agent using the Graph API
Supports any LangChain-compatible chat model (OpenAI, Anthropic, Google, etc.)
Creates a calculator/math agent example that you can run locally
Sets up a new isolated project directory to avoid polluting existing code
Configures environment variables for API keys securely

How to install langgraph-typescript-quickstart

npx skills add https://github.com/langchain-ai/langchain-skills --skill langgraph-typescript-quickstart

Prerequisites

Node.js and npm installed
API key for your chosen LLM provider (OpenAI, Anthropic, Google, etc.)
Basic familiarity with TypeScript

Claude Code

Cursor

Windsurf

Cline

How to use langgraph-typescript-quickstart

1.Run the install command to add the skill
2.Choose your LLM provider and model (e.g., anthropic:claude-sonnet-5)
3.Provide your API key in the generated .env file
4.Run the example agent (e.g., ask it to add 3 and 4)
5.Review the output and refer to langgraph-fundamentals for next steps

Use cases

Good for

Quickly prototype a LangGraph agent locally without complex setup
Learn LangGraph fundamentals with a working calculator example
Test LangGraph with your preferred LLM provider
Build a foundation for more advanced agent workflows

Who it's for

TypeScript developers new to LangGraph
Teams evaluating LangGraph for agent development
Developers wanting to test multi-step reasoning with tools

langgraph-typescript-quickstart FAQ

Which LLM provider should I use?

Any LangChain-compatible model works. The default suggestion is anthropic:claude-sonnet-5, but you can use openai:gpt-4, google-genai:gemini-2.5-flash-lite, or others. Pass it as provider:model.

Do I need LangSmith or Tavily?

No. This quickstart keeps setup minimal with only the API key required. LangSmith and Tavily are optional for advanced features.

Where does the code run?

In a new isolated directory (e.g., langgraph-agent/) created by the skill, keeping your existing project clean.

What happens after the quickstart?

The skill runs one example and stops. Refer to langgraph-fundamentals for deeper learning or use LangChain createAgent for higher-level agent APIs.

Can I use the Functional API instead of Graph API?

The skill defaults to Graph API per the official quickstart, but you can modify the generated code to use Functional API if preferred.

Full instructions (SKILL.md)

Source of truth, from langchain-ai/langchain-skills.

name: langgraph-typescript-quickstart description: "Scaffold a minimal local LangGraph agent in TypeScript by following the official quickstart. Use when the user wants to quickly build or try a LangGraph agent locally."

LangGraph TypeScript quickstart

Follow the live docs — do not invent an alternate API from memory:

https://docs.langchain.com/oss/javascript/langgraph/quickstart

Fetch that page (Docs MCP or HTTP) and implement what it shows (calculator / math agent with the Graph API). Prefer the Graph API path over the Functional API unless the user asks otherwise. Skip graph visualization.

Local setup constraints

Apply these on top of the quickstart (they keep setup minimal and model-agnostic):

Ask which provider/model to use. Showcase that LangGraph works with any LangChain chat model. Suggested prompt:

Which model should this agent use? Pass a provider:model string — e.g. openai:gpt-5.5, anthropic:claude-sonnet-5, google-genai:gemini-2.5-flash-lite. Default if you're unsure: anthropic:claude-sonnet-5.

The docs often hardcode Anthropic — replace with initChatModel("<MODEL>") (or equivalent) using their choice. If using Claude Sonnet 5+, omit temperature / top_p / top_k (unsupported).
Create a new directory (e.g. langgraph-agent/) and do all work there — do not pollute the open project.
Only secret: the provider API key in .env (gitignored). No LangSmith / Tavily unless they ask. Prefer they edit .env themselves — don't paste keys into chat.
Install packages from the quickstart plus the provider package for their model.
Run the example (e.g. “Add 3 and 4.”), show output, then stop. Point to langgraph-fundamentals for next steps. For a higher-level agent API, use LangChain createAgent instead.

Related skills

More from langchain-ai/langchain-skills and the wider catalog.

Official

langchain-ai/langsmith-skills

Build and run evaluation pipelines for LangSmith with LLM judges and custom code evaluators.

4.6k installs