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

deepagents-typescript-quickstart

Scaffold a minimal local Deep Agent in TypeScript using provider-native web search.

1.2k|90|Updated Jan 22, 2026
Search Tags:#web search#Langchain#AI Agent#Quickstart#typeScript#Deep Agents#Scaffold
One-click install
npx skills add https://github.com/langchain-ai/langchain-skills --skill deepagents-typescript-quickstart
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: deepagents-typescript-quickstart
Source: https://github.com/langchain-ai/langchain-skills/tree/main/config/skills/deepagents-typescript-quickstart
Command: npx skills add https://github.com/langchain-ai/langchain-skills --skill deepagents-typescript-quickstart

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Setting up a Deep Agent in TypeScript from scratch involves picking a model provider, wiring a search tool, and following the official quickstart correctly, which is easy to get wrong from memory. This Skill walks through the official quickstart while replacing the default Tavily search dependency with the chosen provider's built-in web search.

Core Features & Use Cases

  • Guided scaffolding: Follows the live LangChain Deep Agents quickstart docs to create a research agent with createDeepAgent, a research system prompt, and an invoke call.
  • Provider-native search: Replaces Tavily with built-in web search from Anthropic, OpenAI, or Google, so only one provider API key is needed.
  • Clean local setup: Creates a new isolated directory, installs only the required packages, and keeps secrets in a gitignored .env file.
  • Use Case: A developer wants to try Deep Agents locally with their Anthropic API key; the Skill scaffolds a deep-agent/ project, runs a research question like "What is LangGraph?", and shows the output.

Quick Start

Use the deepagents-typescript-quickstart skill to scaffold a local Deep Agent in TypeScript with Anthropic's built-in web search and run the research example.

Frequently Asked Questions about deepagents-typescript-quickstart

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

FAQPage Schema
How do I build a Deep Agent in TypeScript locally?▼

Follow the official LangChain Deep Agents quickstart, which uses createDeepAgent with a research system prompt and an invoke call. This Skill scaffolds the project in a new directory, installs the required packages, and runs a research example. Node 22 or later is required.

How to use Deep Agents without Tavily search?▼

Replace the Tavily search tool with your model provider's built-in web search. Anthropic offers webSearch tools via @langchain/anthropic, OpenAI accepts a web_search tool type, and Google supports a google_search tool, so no separate search API key is needed.

Which model providers work with Deep Agents TypeScript?▼

Deep Agents are model-agnostic and accept a provider:model string such as anthropic:claude-sonnet-5, openai:gpt-5.5, or google-genai:gemini-3.5-flash. Anthropic, OpenAI, and Google are preferred because they offer built-in web search tools.

What Node version does the Deep Agents quickstart require?▼

The Deep Agents TypeScript quickstart requires Node 22 or later. You also need an API key for your chosen model provider, stored in a gitignored .env file inside the new project directory.

Does the Deep Agents quickstart need LangSmith tracing?▼

No, LangSmith tracing is optional and skipped by default in this setup. The only required secret is the model provider's API key. You can enable tracing later if you want observability into agent runs.

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
bosens-Chinabosens-China
10

deep-agents-orchestration

Orchestrate subagents, plan tasks, and manage human approvals in Deep Agents.

Community
Advanced
artgaurav16420-ossartgaurav16420-oss

langsmith-fetch

Fetch and analyze recent LangSmith traces to debug LangChain and LangGraph agents.

Community
Intermediate
coinvest518coinvest518

langgraph-docs

Fetch and reference LangGraph Python documentation for building stateful agents.

Community
Intermediate