Agent Design

Agent Design

> Design LLM agents with memory and tools.

Quick Start

from langchain_openai import ChatOpenAI
from langchain.agents import create_tool_calling_agent, AgentExecutor
from langchain_core.tools import tool
from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder

@tool
def search(query: str) -> str:
    """Search the web for information."""
    return f"Results for: {query}"

tools = [search]

prompt = ChatPromptTemplate.from_messages([
    ("system", "You are a helpful assistant."),
    ("human", "{input}"),
    MessagesPlaceholder("agent_scratchpad"),
])

llm = ChatOpenAI(model="gpt-4o", temperature=0)
agent = create_tool_calling_agent(llm, tools, prompt)
executor = AgentExecutor(agent=agent, tools=tools, verbose=True)
executor.invoke({"input": "Find information about Python"})

When to Use

• ✅ Autonomous AI assistants
• ✅ Need reasoning and tool usage
• ❌ Not for simple Q&A tasks

Step-by-Step Instructions

1. Define tools (search, calculator, API)
2. Setup system prompt with agent role
3. Add memory (buffer, summary)
4. Test reasoning cycle

Dependencies

pip install langchain langchain-openai langchain-core

Examples

Input: "What's the weather in Moscow?" → Output: Agent calls weather API, returns answer

Resources

• LangChain Agents
• [Examples](./examples/)

Validation

1. Agent selects correct tools
2. Memory preserves dialogue context
3. Reasoning chain is logical and complete

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: ssrjkk
• Source: ssrjkk/claude-skills
• License: MIT
• Homepage: https://claude.ai

Install and usage instructions live in the source repository linked above.