LangChain DeepAgents

DeepAgents is a LangChain-based framework for building autonomous AI agents that plan, research, critique, and iterate. Muster captures DeepAgents executions through the standard LangChain CallbackHandler.

Setup

1. Install dependencies

%pip install langfuse deepagents tavily-python

2. Configure credentials

import os

os.environ["LANGFUSE_PUBLIC_KEY"] = "pk-lf-..."
os.environ["LANGFUSE_SECRET_KEY"] = "sk-lf-..."
os.environ["LANGFUSE_HOST"] = "https://app.getmuster.io"
os.environ["ANTHROPIC_API_KEY"] = "sk-..."
os.environ["TAVILY_API_KEY"] = "tvly-..."

3. Initialize the callback handler

from langfuse import get_client
from langfuse.langchain import CallbackHandler

langfuse = get_client()

if langfuse.auth_check():
    print("Muster client is authenticated and ready!")

langfuse_handler = CallbackHandler()

Example 1: simple research agent

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,
):
    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="You are an expert researcher.",
)

result = agent.invoke(
    {"messages": [{"role": "user", "content": "What is Muster and how does it help with LLM observability?"}]},
    config={"callbacks": [langfuse_handler]},
)

DeepAgents simple-research trace

Example 2: multi-agent with sub-agents

from deepagents import create_deep_agent

research_sub_agent = {
    "name": "research-agent",
    "description": "Used to research more in-depth questions.",
    "system_prompt": "You are a dedicated researcher.",
    "tools": [internet_search],
}

critique_sub_agent = {
    "name": "critique-agent",
    "description": "Critiques the final report.",
    "system_prompt": "You are a dedicated editor.",
}

agent = create_deep_agent(
    tools=[internet_search],
    system_prompt="You are an expert researcher.",
    subagents=[critique_sub_agent, research_sub_agent],
)

result = agent.invoke(
    {"messages": [{"role": "user", "content": "What is Muster?"}]},
    config={"callbacks": [langfuse_handler]},
)

DeepAgents multi-agent trace

The trace shows agent planning, sub-agent interactions, tool invocations (web searches), full conversation history, and per-step latencies.