langchain-typescript-quickstart
langchain-ai/langchain-skills
Scaffold a minimal LangChain TypeScript agent locally following the official quickstart.
What is langchain-typescript-quickstart?
This skill sets up a working LangChain agent in TypeScript by implementing the official quickstart guide. Use it when you need to quickly build or test a LangChain agent locally with minimal setup.
- Creates a new isolated directory for the LangChain project
- Prompts you to select a model provider (OpenAI, Anthropic, Google, etc.)
- Installs required dependencies and provider packages
- Configures environment variables for API authentication
- Implements a weather agent example from the official docs
- Runs the agent and displays output
How to install langchain-typescript-quickstart
npx skills add https://github.com/langchain-ai/langchain-skills --skill langchain-typescript-quickstart- Node.js 22 or higher
- API key for your chosen model provider (OpenAI, Anthropic, Google, etc.)
How to use langchain-typescript-quickstart
- 1.Run the skill to create a new langchain-agent directory
- 2.Select your preferred model provider when prompted (default: anthropic:claude-sonnet-5)
- 3.Add your provider API key to the generated .env file
- 4.The skill installs dependencies and runs the weather agent example
- 5.Review the output and explore the generated code for next steps
Use cases
- Quickly prototype a LangChain agent without manual setup
- Test LangChain functionality with your preferred model provider
- Learn LangChain basics by running a working example
- Set up a foundation for building more complex agents
- TypeScript developers new to LangChain
- Engineers prototyping AI agents
- Teams evaluating LangChain for production use
langchain-typescript-quickstart FAQ
No. This quickstart uses only a model provider API key. LangSmith and Tavily are optional and only needed if you explicitly request them.
Yes. The skill prompts you to choose any provider:model combination (e.g., openai:gpt-4, google-genai:gemini-2.5-flash-lite). LangChain is model-agnostic.
The skill stops after showing output. Refer to the langchain-fundamentals skill for guidance on building more complex agents.
No. The skill creates a new isolated directory (langchain-agent/) and does not modify your open project.
Full instructions (SKILL.md)
Source of truth, from langchain-ai/langchain-skills.
name: langchain-typescript-quickstart description: "Scaffold a minimal local LangChain agent in TypeScript by following the official quickstart. Use when the user wants to quickly build or try a LangChain agent locally."
LangChain TypeScript quickstart
Follow the live docs — do not invent an alternate API from memory:
https://docs.langchain.com/oss/javascript/langchain/quickstart
Fetch that page (Docs MCP or HTTP) and implement what it shows (weather agent + createAgent). Requires Node 22+.
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 LangChain is model-agnostic. Suggested prompt:
Which model should this agent use? Pass a
provider:modelstring — 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.Swap the quickstart's model string for their choice (or the default).
-
Create a new directory (e.g.
langchain-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.envthemselves — don't paste keys into chat. -
Install the provider package needed for their model if the quickstart's base install isn't enough.
-
Run the example, show output, then stop. Point to
langchain-fundamentalsfor next steps.
Related skills
More from langchain-ai/langchain-skills and the wider catalog.
langgraph-cli
Scaffold, develop, build, and deploy LangGraph applications from CLI.
langgraph-fundamentals
Master LangGraph's directed graph workflows for building stateful, multi-step agent applications.
langgraph-human-in-the-loop
Pause LangGraph execution for human approval, validation, or input—then resume with their decision.
langgraph-persistence
Persist LangGraph state with checkpointers, thread IDs, and time travel for durable multi-turn conversations.
langgraph-python-quickstart
Scaffold a minimal LangGraph agent in Python following the official quickstart.
langgraph-typescript-quickstart
Scaffold a minimal LangGraph agent in TypeScript from the official quickstart.