deepagents-python-quickstart

1# Deep Agents Python quickstart
2 
3Follow the live docs — do not invent an alternate API from memory:
4 
5**https://docs.langchain.com/oss/python/deepagents/quickstart**
6 
7Fetch 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?”).
8 
9## Local setup constraints
10 
11Apply these on top of the quickstart (they keep setup minimal and model-agnostic):
12 
131. **Ask** which provider/model to use. Showcase that Deep Agents are model-agnostic. Suggested prompt:
14 
15   > 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`**.  
16   > We'll use that provider's built-in web search (no separate search API key).
17 
182. Create a **new** directory (e.g. `deep-agent/`) and do all work there — do not pollute the open project.
19 
203. **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):
21 
22   | Provider | Built-in search tool |
23   |----------|----------------------|
24   | Anthropic | `{"type": "web_search_20260209", "name": "web_search", "max_uses": 5}` |
25   | OpenAI | `{"type": "web_search"}` |
26…