Langchain – Agents

LangChain – Agents

(1) What Is An Agent ?

from langchain.agents import create_agent
agent = create_agent(
        model = "ollama:qwen3:0.6b",
        tools= tools,
        system_prompt="Your System Prompt"
        )

(2) Core Components Of Agent

(3) Model Component

from langchain.agents import create_agent
agent = create_agent(
        model="qwen3:0.6b",
        tools=[tools],
        system_prompt = "System Prompt"
        )

(4) Tools Component

from langchain.agents import create_agent
from langchain.tools import tools

@tool 
def search(query:str)->str:
    """Search For Information"""
    return f"Result For {query}"
    

agent = create_agent(
        model="ollama:qwen3:0.6b",
        tools = [search],
        system_prompt = "You Are An Healpful AI Assistant"
        )

(5) System Prompt

agent = create_agent(
        model = "ollama:qwen3:0.6b",
        tools = [tools],
        system_prompt = "You Will Answar My Questions"
        )

(6) Structured Output

from pydentic import BaseModel
from langchain.agents import create_agents

class Answer(BaseModel):
      summary:str
      confidence:float
     
agent = create_agent(
        model = "ollama:qwen3:0.6b",
        tools = [tools],
        system_prompt = "System Prompt",
        response_format = Answer
        )

(7) Agent Invocation

from langchain.agents create_agent

from langchain_core.utils.uuid import uuid7
from langgraph.checkpoint.memory import InMemorySaver

agent = create_agent(
        model="ollama:qwen3:0.6m",
        tools=[],
        checkpointer = InMemorySaver(),
        )

config = {"configurable": {"thread_id": str(uuid7())}}


result = agent.invoke({
    "messages": [{"role": "user", "content":"What Is The Weather In Bhubaneswar"}],
    config = config,
})

# A Next Conversation Uses The Same Thread_Id 

result = agent.invoke({"messages":[{"role":"user", "content":"What About Tomorrow"}]},
    config = config
    )