The World of
AI Agents

copy # Install the required packages pip install anthropic langchain openai python-dotenv # Create a .env file with your API key ANTHROPIC_API_KEY=sk-ant-...your-key-here... # Install dotenv to load environment variables from dotenv import load_dotenv load_dotenv()

copy import anthropic import json, requests # Define tools as a list of JSON schemas tools = [ { "name": "web_search", "description": "Search the web for current information about any topic", "input_schema": { "type": "object", "properties": { "query": {"type": "string", "description": "The search query"} }, "required": ["query"] } }, { "name": "run_python", "description": "Execute Python code and return the output", "input_schema": { "type": "object", "properties": { "code": {"type": "string", "description": "Python code to execute"} }, "required": ["code"] } } ]

copy def execute_tool(tool_name, tool_input): if tool_name == "web_search": # Connect to your search API (Serper, Tavily, etc.) query = tool_input["query"] response = requests.get( "https://api.tavily.com/search", json={"query": query, "api_key": "YOUR_KEY"} ) return response.json()["results"][:3] elif tool_name == "run_python": import io, contextlib output = io.StringIO() with contextlib.redirect_stdout(output): exec(tool_input["code"]) return output.getvalue() return "Tool not found"

copy client = anthropic.Anthropic() def run_agent(user_goal): messages = [{"role": "user", "content": user_goal}] while True: # Ask the LLM what to do next response = client.messages.create( model="claude-sonnet-4-20250514", max_tokens=4096, tools=tools, messages=messages, system="You are a helpful agent. Think step by step." ) # Add assistant's response to history messages.append({"role": "assistant", "content": response.content}) # If done, return the final answer if response.stop_reason == "end_turn": return response.content[0].text # Execute each tool call tool_results = [] for block in response.content: if block.type == "tool_use": print(f"🛠️ Calling: {block.name}") result = execute_tool(block.name, block.input) tool_results.append({ "type": "tool_result", "tool_use_id": block.id, "content": str(result) }) # Feed results back to agent messages.append({"role": "user", "content": tool_results}) # Run your agent! result = run_agent("Research the top 3 AI frameworks in 2025 and compare them") print(result)

copy # Simple long-term memory with ChromaDB (vector DB) import chromadb chroma = chromadb.Client() memory = chroma.create_collection("agent_memory") def remember(text, session_id): memory.add(documents=[text], ids=[session_id]) def recall(query, n=3): results = memory.query(query_texts=[query], n_results=n) return results["documents"][0] # Deploy as FastAPI endpoint from fastapi import FastAPI app = FastAPI() @app.post("/agent") async def agent_endpoint(goal: str): result = run_agent(goal) remember(result, goal[:50]) return {"result": result}