mcpField
D
deepagents-python-quickstart

deepagents-python-quickstart

github.com/langchain-aiAgent workflows1.3k5.2k installsUpdated 3d ago
Repository

About deepagents-python-quickstart

A skill for developers building a research agent with the official Deep Agents Python quickstart. It uses the selected provider's built-in web search and keeps setup and execution in a new project directory.

Key features

  1. 01Follow the official quickstart
  2. 02Build a research agent with create_deep_agent
  3. 03Choose a model provider and built-in web search
  4. 04Install dependencies in a separate directory
  5. 05Run the research example and show output

Use cases

  1. 01Create a research-focused Python agent
  2. 02Use a provider's built-in web search
  3. 03Run a research question and inspect the result

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/langchain-ai/langchain-skills --skill deepagents-python-quickstart

After installing, type/deepagents-python-quickstartin Claude Code to confirm it loaded.

Details

From the project README

Deep Agents Python quickstart

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

https://docs.langchain.com/oss/python/deepagents/quickstart

Fetch that page (Docs MCP or HTTP) and implement the research-agent shape it shows (create_deep_agent, research system prompt, invoke with a research question like “What is LangGraph?”).

Local setup constraints

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

  1. Ask which provider/model to use. Showcase that Deep Agents are model-agnostic. 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-3.5-flash. Default if you're unsure: anthropic:claude-sonnet-5.
    We'll use that provider's built-in web search (no separate search API key).

  2. Create a new directory (e.g. deep-agent/) and do all work there — do not pollute the open project.

  3. Do not use Tavily (or any second search vendor). Replace the quickstart's internet_search / Tavily tool with the chosen provider's built-in web search. Look up the current tool shape on that provider's LangChain chat docs (examples as of writing — re-check if needed):

    ProviderBuilt-in search tool
    Anthropic{"type": "web_search_20260209", "name": "web_search", "max_uses": 5}
    OpenAI{"type": "web_search"}
    Google{"google_search": {}}

    Prefer Anthropic / OpenAI / Google so provider search is available. Only secret: that provider's API key in .env (gitignored). Skip LangSmith tracing unless they ask.

  4. Install deepagents (+ python-dotenv) and the provider package for their model — not tavily-python.

  5. Run the research example, show output, then stop. Point to deep-agents-core / customization / Managed Deep Agents for next steps.

Comments

Markdown supported