Customize Deep Agents

Learn how to customize deep agents with system prompts, tools, subagents, and more

create_deep_agent has the following core configuration options:

• Model
• Tools
• System Prompt

And the following configurable features:

• Middleware
• Subagents
• Filesystem backend
• Human-in-the-loop
• Skills
• Memory

Core configuration

create_deep_agent(
    name: str | None = None,
    model: str | BaseChatModel | None = None,
    tools: Sequence[BaseTool | Callable | dict[str, Any]] | None = None,
    *,
    system_prompt: str | SystemMessage | None = None
) -> CompiledStateGraph

For more information, see create_deep_agent.

Model

By default, deepagents uses claude-sonnet-4-5-20250929. You can customize the model by passing any supported model identifier string or LangChain model object.

Use the provider:model format (for example openai:gpt-5) to quickly switch between models.

OpenAI

👉 Read the OpenAI chat model integration docs

pip install -U "langchain[openai]"

init_chat_model

import os
from langchain.chat_models import init_chat_model
from deepagents import create_deep_agent

os.environ["OPENAI_API_KEY"] = "sk-..."

model = init_chat_model(model="openai:gpt-4.1")
agent = create_deep_agent(model=model)

import os
from langchain_openai import ChatOpenAI
from langchain.chat_models import init_chat_model
from deepagents import create_deep_agent

os.environ["OPENAI_API_KEY"] = "sk-..."

model = init_chat_model(
    model=ChatOpenAI(model="gpt-4.1")
)
agent = create_deep_agent(model=model)

Anthropic

👉 Read the Anthropic chat model integration docs

pip install -U "langchain[anthropic]"

init_chat_model

import os
from langchain.chat_models import init_chat_model
from deepagents import create_deep_agent

os.environ["ANTHROPIC_API_KEY"] = "sk-..."

model = init_chat_model(model="claude-sonnet-4-5-20250929")
agent = create_deep_agent(model=model)

import os
from langchain_anthropic import ChatAnthropic
from langchain.chat_models import init_chat_model
from deepagents import create_deep_agent

os.environ["ANTHROPIC_API_KEY"] = "sk-..."

model = init_chat_model(
    model=ChatAnthropic(model="claude-sonnet-4-5-20250929")
)
agent = create_deep_agent(model=model)

Azure

👉 Read the Azure chat model integration docs

pip install -U "langchain[openai]"

init_chat_model

import os
from langchain.chat_models import init_chat_model
from deepagents import create_deep_agent

os.environ["AZURE_OPENAI_API_KEY"] = "..."
os.environ["AZURE_OPENAI_ENDPOINT"] = "..."
os.environ["OPENAI_API_VERSION"] = "2025-03-01-preview"

model = init_chat_model(
    model="azure_openai:gpt-4.1",
    azure_deployment=os.environ["AZURE_OPENAI_DEPLOYMENT_NAME"],
)
agent = create_deep_agent(model=model)

import os
from langchain_openai import AzureChatOpenAI
from langchain.chat_models import init_chat_model
from deepagents import create_deep_agent

os.environ["AZURE_OPENAI_API_KEY"] = "..."
os.environ["AZURE_OPENAI_ENDPOINT"] = "..."
os.environ["OPENAI_API_VERSION"] = "2025-03-01-preview"

model = init_chat_model(
    model=AzureChatOpenAI(
        model="gpt-4.1",
        azure_deployment=os.environ["AZURE_OPENAI_DEPLOYMENT_NAME"]
    )
)
agent = create_deep_agent(model=model)

Google Gemini

👉 Read the Google GenAI chat model integration docs

pip install -U "langchain[google-genai]"

init_chat_model

import os
from langchain.chat_models import init_chat_model
from deepagents import create_deep_agent

os.environ["GOOGLE_API_KEY"] = "..."

model = init_chat_model(model="google_genai:gemini-2.5-flash-lite")
agent = create_deep_agent(model=model)

import os
from langchain_google_genai import ChatGoogleGenerativeAI
from langchain.chat_models import init_chat_model
from deepagents import create_deep_agent

os.environ["GOOGLE_API_KEY"] = "..."

model = init_chat_model(
    model=ChatGoogleGenerativeAI(model="gemini-2.5-flash-lite")
)
agent = create_deep_agent(model=model)

AWS Bedrock

👉 Read the AWS Bedrock chat model integration docs

pip install -U "langchain[aws]"

init_chat_model

from langchain.chat_models import init_chat_model
from deepagents import create_deep_agent

# Follow the steps here to configure your credentials:
# https://docs.aws.amazon.com/bedrock/latest/userguide/getting-started.html

model = init_chat_model(
    model="anthropic.claude-3-5-sonnet-20240620-v1:0",
    model_provider="bedrock_converse",
)
agent = create_deep_agent(model=model)

from langchain_aws import ChatBedrock
from langchain.chat_models import init_chat_model
from deepagents import create_deep_agent

# Follow the steps here to configure your credentials:
# https://docs.aws.amazon.com/bedrock/latest/userguide/getting-started.html

model = init_chat_model(
    model=ChatBedrock(model="anthropic.claude-3-5-sonnet-20240620-v1:0")
)
agent = create_deep_agent(model=model)

HuggingFace

👉 Read the HuggingFace chat model integration docs

pip install -U "langchain[huggingface]"

init_chat_model

import os
from langchain.chat_models import init_chat_model
from deepagents import create_deep_agent

os.environ["HUGGINGFACEHUB_API_TOKEN"] = "hf_..."

model = init_chat_model(
    model="microsoft/Phi-3-mini-4k-instruct",
    model_provider="huggingface",
    temperature=0.7,
    max_tokens=1024,
)
agent = create_deep_agent(model=model)

import os
from langchain_huggingface import ChatHuggingFace, HuggingFaceEndpoint
from langchain.chat_models import init_chat_model
from deepagents import create_deep_agent

os.environ["HUGGINGFACEHUB_API_TOKEN"] = "hf_..."

llm = HuggingFaceEndpoint(
    repo_id="microsoft/Phi-3-mini-4k-instruct",
    temperature=0.7,
    max_length=1024,
)
model = init_chat_model(
    model=ChatHuggingFace(llm=llm)
)
agent = create_deep_agent(model=model)

Tools

In addition to built-in tools for planning, file management, and subagent spawning, you can provide custom tools:

import os
from typing import Literal
from tavily import TavilyClient
from deepagents import create_deep_agent

tavily_client = TavilyClient(api_key=os.environ["TAVILY_API_KEY"])

def internet_search(
    query: str,
    max_results: int = 5,
    topic: Literal["general", "news", "finance"] = "general",
    include_raw_content: bool = False,
):
    """Run a web search"""
    return tavily_client.search(
        query,
        max_results=max_results,
        include_raw_content=include_raw_content,
        topic=topic,
    )

agent = create_deep_agent(
    tools=[internet_search]
)

System prompt

Deep agents come with a built-in system prompt inspired by Claude Code’s system prompt. The default system prompt contains detailed instructions for using the built-in planning tool, file system tools, and subagents.

Each deep agent should include a custom system prompt specific to its specific use case.

from deepagents import create_deep_agent

research_instructions = """\
You are an expert researcher. Your job is to conduct \
thorough research, and then write a polished report. \
"""

agent = create_deep_agent(
    system_prompt=research_instructions,
)

Features

Middleware

Middleware provides a way to more tightly control what happens inside an agent. You can provide additional middleware to extend functionality, add tools, or implement custom hooks:

from langchain.tools import tool
from langchain.agents.middleware import wrap_tool_call
from deepagents import create_deep_agent


@tool
def get_weather(city: str) -> str:
    """Get the weather in a city."""
    return f"The weather in {city} is sunny."


call_count = [0]  # Use list to allow modification in nested function

@wrap_tool_call
def log_tool_calls(request, handler):
    """Intercept and log every tool call - demonstrates cross-cutting concern."""
    call_count[0] += 1
    tool_name = request.name if hasattr(request, 'name') else str(request)

    print(f"[Middleware] Tool call #{call_count[0]}: {tool_name}")
    print(f"[Middleware] Arguments: {request.args if hasattr(request, 'args') else 'N/A'}")

    # Execute the tool call
    result = handler(request)

    # Log the result
    print(f"[Middleware] Tool call #{call_count[0]} completed")

    return result


agent = create_deep_agent(
    tools=[get_weather],
    middleware=[log_tool_calls],
)

Subagents

To isolate detailed work and avoid context bloat, use subagents:

import os
from typing import Literal
from tavily import TavilyClient
from deepagents import create_deep_agent

tavily_client = TavilyClient(api_key=os.environ["TAVILY_API_KEY"])

def internet_search(
    query: str,
    max_results: int = 5,
    topic: Literal["general", "news", "finance"] = "general",
    include_raw_content: bool = False,
):
    """Run a web search"""
    return tavily_client.search(
        query,
        max_results=max_results,
        include_raw_content=include_raw_content,
        topic=topic,
    )

research_subagent = {
    "name": "research-agent",
    "description": "Used to research more in depth questions",
    "system_prompt": "You are a great researcher",
    "tools": [internet_search],
    "model": "openai:gpt-4.1",  # Optional override, defaults to main agent model
}
subagents = [research_subagent]

agent = create_deep_agent(
    model="claude-sonnet-4-5-20250929",
    subagents=subagents
)

For more information, see Subagents.

Filesystem backend

For skills and memory, you must provide a filesystem backend. You may also find it helpful to use the backend with custom tools.

StateBackend

An ephemeral filesystem backend stored in langgraph state. This filesystem only persists for a single thread.

# By default we provide a StateBackend
agent = create_deep_agent()

# Under the hood, it looks like
from deepagents.backends import StateBackend

agent = create_deep_agent(
    backend=(lambda rt: StateBackend(rt))   # Note that the tools access State through the runtime.state
)

FilesystemBackend

The local machine’s filesystem.

from deepagents.backends import FilesystemBackend

agent = create_deep_agent(
    backend=FilesystemBackend(root_dir=".", virtual_mode=True)
)

StoreBackend

A filesystem that provides long-term storage that is persisted across threads.

from langgraph.store.memory import InMemoryStore
from deepagents.backends import StoreBackend

agent = create_deep_agent(
    backend=(lambda rt: StoreBackend(rt)),   # Note that the tools access Store through the runtime.store
    store=InMemoryStore()
)

CompositeBackend

A flexible backen where you can specify different routes in the filesystem to point towards different backends.

from deepagents import create_deep_agent
from deepagents.backends import CompositeBackend, StateBackend, StoreBackend
from langgraph.store.memory import InMemoryStore

composite_backend = lambda rt: CompositeBackend(
    default=StateBackend(rt),
    routes={
        "/memories/": StoreBackend(rt),
    }
)

agent = create_deep_agent(
    backend=composite_backend,
    store=InMemoryStore()  # Store passed to create_deep_agent, not backend
)

For more information, see Backends.

Human-in-the-loop

Some tool operations may be sensitive and require human approval before execution. You can configure the approval for each tool:

from langchain.tools import tool
from deepagents import create_deep_agent
from langgraph.checkpoint.memory import MemorySaver

@tool
def delete_file(path: str) -> str:
    """Delete a file from the filesystem."""
    return f"Deleted {path}"

@tool
def read_file(path: str) -> str:
    """Read a file from the filesystem."""
    return f"Contents of {path}"

@tool
def send_email(to: str, subject: str, body: str) -> str:
    """Send an email."""
    return f"Sent email to {to}"

# Checkpointer is REQUIRED for human-in-the-loop
checkpointer = MemorySaver()

agent = create_deep_agent(
    model="claude-sonnet-4-5-20250929",
    tools=[delete_file, read_file, send_email],
    interrupt_on={
        "delete_file": True,  # Default: approve, edit, reject
        "read_file": False,   # No interrupts needed
        "send_email": {"allowed_decisions": ["approve", "reject"]},  # No editing
    },
    checkpointer=checkpointer  # Required!
)

You can configure interrupt for agents and subagents on tool call as well as from within tool calls. For more information, see Human-in-the-loop.

Skills

You can use skills to provide your deep agent with new capabilities and expertise. While tools tend to cover lower level functionality like native file system actions or planning, skills can contain detailed instructions on how to complete tasks, reference info, and other assets, such as templates. These files are only loaded by the agent when the agent has determined that the skill is useful for the current prompt. This progressive disclosure reduces the amount of tokens and context the agent has to consider upon startup.

For example skills, see Deep Agent example skills.

To add skills to your deep agent, pass them as an argument to create_deep_agent:

StateBackend

from urllib.request import urlopen
from deepagents import create_deep_agent
from langgraph.checkpoint.memory import MemorySaver

checkpointer = MemorySaver()

skill_url = "https://raw.githubusercontent.com/langchain-ai/deepagentsjs/refs/heads/main/examples/skills/langgraph-docs/SKILL.md"
with urlopen(skill_url) as response:
    skill_content = response.read().decode('utf-8')

skills_files = {
    "/skills/langgraph-docs/SKILL.md": skill_content
}

agent = create_deep_agent(
    skills=["./skills/"],
    checkpointer=checkpointer,
)

result = agent.invoke(
    {
        "messages": [
            {
                "role": "user",
                "content": "What is langgraph?",
            }
        ],
        # Seed the default StateBackend's in-state filesystem (virtual paths must start with "/").
        "files": skills_files
    },
    config={"configurable": {"thread_id": "12345"}},
)

StoreBackend

from urllib.request import urlopen
from deepagents import create_deep_agent
from deepagents.backends import StoreBackend
from langgraph.store.memory import InMemoryStore


store = InMemoryStore()

skill_url = "https://raw.githubusercontent.com/langchain-ai/deepagentsjs/refs/heads/main/examples/skills/langgraph-docs/SKILL.md"
with urlopen(skill_url) as response:
    skill_content = response.read().decode('utf-8')

store.put(
    namespace=("filesystem",),
    key="/skills/langgraph-docs/SKILL.md",
    value=skill_content
)

agent = create_deep_agent(
    backend=(lambda rt: StoreBackend(rt)),
    store=store,
    skills=["./skills/"]
)

result = agent.invoke(
    {
        "messages": [
            {
                "role": "user",
                "content": "What is langgraph?",
            }
        ]
    },
    config={"configurable": {"thread_id": "12345"}},
)

FilesystemBackend

from deepagents import create_deep_agent
from langgraph.checkpoint.memory import MemorySaver
from deepagents.backends.filesystem import FilesystemBackend

# Checkpointer is REQUIRED for human-in-the-loop
checkpointer = MemorySaver()

agent = create_deep_agent(
    backend=FilesystemBackend(root_dir="/Users/user/{project}"),
    skills=["/Users/user/{project}/skills/"],
    interrupt_on={
        "write_file": True,  # Default: approve, edit, reject
        "read_file": False,  # No interrupts needed
        "edit_file": True    # Default: approve, edit, reject
    },
    checkpointer=checkpointer,  # Required!
)

result = agent.invoke(
    {
        "messages": [
            {
                "role": "user",
                "content": "What is langgraph?",
            }
        ]
    },
    config={"configurable": {"thread_id": "12345"}},
)

Memory

Use AGENTS.md files to provide extra context to your deep agent.

You can pass one or more file paths to the memory parameter when creating your deep agent:

StateBackend

from urllib.request import urlopen

from deepagents import create_deep_agent
from deepagents.backends.utils import create_file_data
from langgraph.checkpoint.memory import MemorySaver

with urlopen("https://raw.githubusercontent.com/langchain-ai/deepagents/refs/heads/master/examples/text-to-sql-agent/AGENTS.md") as response:
    agents_md = response.read().decode("utf-8")
checkpointer = MemorySaver()

agent = create_deep_agent(
    memory=[
        "/AGENTS.md"
    ],
    checkpointer=checkpointer,
)

result = agent.invoke(
    {
        "messages": [
            {
                "role": "user",
                "content": "Please tell me what's in your memory files.",
            }
        ],
        # Seed the default StateBackend's in-state filesystem (virtual paths must start with "/").
        "files": {"/AGENTS.md": create_file_data(agents_md)},
    },
    config={"configurable": {"thread_id": "123456"}},
)

StoreBackend

from urllib.request import urlopen

from deepagents import create_deep_agent
from deepagents.backends import StoreBackend
from deepagents.backends.utils import create_file_data
from langgraph.store.memory import InMemoryStore

with urlopen("https://raw.githubusercontent.com/langchain-ai/deepagents/refs/heads/master/examples/text-to-sql-agent/AGENTS.md") as response:
    agents_md = response.read().decode("utf-8")

# Create the store and add the file to it
store = InMemoryStore()
file_data = create_file_data(agents_md)
store.put(
    namespace=("filesystem",),
    key="/AGENTS.md",
    value=file_data
)

agent = create_deep_agent(
    backend=(lambda rt: StoreBackend(rt)),
    store=store,
    memory=[
        "/AGENTS.md"
    ]
)

result = agent.invoke(
    {
        "messages": [
            {
                "role": "user",
                "content": "Please tell me what's in your memory files.",
            }
        ],
        "files": {"/AGENTS.md": create_file_data(agents_md)},
    },
    config={"configurable": {"thread_id": "12345"}},
)

FilesystemBackend

from deepagents import create_deep_agent
from langgraph.checkpoint.memory import MemorySaver
from deepagents.backends import FilesystemBackend

# Checkpointer is REQUIRED for human-in-the-loop
checkpointer = MemorySaver()

agent = create_deep_agent(
    backend=FilesystemBackend(root_dir="/Users/user/{project}"),
    memory=[
        "./AGENTS.md"
    ],
    interrupt_on={
        "write_file": True,  # Default: approve, edit, reject
        "read_file": False,  # No interrupts needed
        "edit_file": True    # Default: approve, edit, reject
    },
    checkpointer=checkpointer,  # Required!
)

---

Edit this page on GitHub or file an issue.

Connect these docs to Claude, VSCode, and more via MCP for real-time answers.