The create_deep_agent function is the main entry point for creating deep agents. It returns a compiled agent graph with built-in planning, file system access, and subagent capabilities.
Learn more: For tutorials and guides on building deep agents, see the Deep Agents documentation.
from deepagents import create_deep_agent
# Create an agent with default settings
agent = create_deep_agent()
# Run the agent
result = agent.invoke({
"messages": [{"role": "user", "content": "Research quantum computing"}]
})from deepagents import create_deep_agent
agent = create_deep_agent(
model="anthropic:claude-sonnet-4-20250514",
tools=[my_search_tool],
subagents=[research_subagent],
system_prompt="You are an expert researcher.",
)