LangGraph.js Quickstart

A minimal LangGraph.js research graph. Clone, set your key, deploy — done.

Deploy in 3 steps

1. Clone & install

git clone https://github.com/Crewship/langgraph-js-quickstart.git
cd langgraph-js-quickstart
# Install Crewship CLI tool
curl -fsSL crewship.dev/install | bash
npm install

2. Set your API key

crewship env set OPENAI_API_KEY=sk-...

Using a different provider? Swap the ChatOpenAI import in src/graph.ts and set the matching key — e.g. ANTHROPIC_API_KEY for Claude, GROQ_API_KEY for Groq (free tier).

3. Deploy

crewship deploy

That's it. Your research graph is live.

Run locally (optional)

cp .env.example .env   # add your OPENAI_API_KEY
npm start

---

What's inside

A two-node graph powered by GPT-4o mini:

• researcher — finds 5 key points about a topic
• reporter — turns them into a short markdown report

Change the topic in src/main.ts.

Customise

• Graph logic → src/graph.ts
• Entry point → src/main.ts

Learn more

• LangGraph.js documentation
• LangChain.js docs