Ollama Local AI Agent Architecture Generator

use_case: Customer service chatbot for healthcare clinic, privacy_level: HIPAA compliant with full data isolation, primary_functions: appointment scheduling, symptom triage, FAQ responses, patient record lookup

**AGENT OVERVIEW** HealthCare Assistant Agent - A HIPAA-compliant local AI agent for healthcare clinics that provides patient support while maintaining complete data privacy. Runs entirely on local infrastructure using Ollama with medical knowledge fine-tuning. **ARCHITECTURE DIAGRAM** ``` ┌─────────────────┐ ┌──────────────┐ ┌─────────────┐ │ Web Interface │────│ Agent Core │────│ Ollama API │ │ (Frontend) │ │ (Controller) │ │ (LLM) │ └─────────────────┘ └──────────────┘ └─────────────┘ │ │ │ │ ┌─────────────────┐ │ └──────────────│ SQLite Database │──────────┘ │ (Patient Data) │ └─────────────────┘ ``` **OLLAMA SETUP** ```bash # Install Ollama curl -fsSL https://ollama.ai/install.sh | sh # Pull base model ollama pull llama2:13b # Create custom healthcare model ollama create healthcare-agent -f ./Modelfile ``` **MODELFILE** ``` FROM llama2:13b PARAMETER temperature 0.3 PARAMETER top_p 0.9 PARAMETER stop "Patient:" PARAMETER stop "Human:" SYSTEM You are a professional healthcare assistant. Always maintain patient confidentiality and provide accurate, helpful responses about appointments and general health inquiries. Never provide medical diagnoses. ``` **IMPLEMENTATION CODE** ```python import asyncio import sqlite3 from fastapi import FastAPI, HTTPException from pydantic import BaseModel import httpx import logging from datetime import datetime class HealthcareAgent: def __init__(self): self.ollama_url = "http://localhost:11434" self.db_path = "healthcare_data.db" self.setup_database() def setup_database(self): """Initialize HIPAA-compliant local database""" conn = sqlite3.connect(self.db_path) conn.execute('''CREATE TABLE IF NOT EXISTS appointments (id INTEGER PRIMARY KEY, patient_id TEXT, datetime TEXT, status TEXT)''') conn.close() async def process_query(self, query: str, patient_id: str): """Process patient query with privacy protection""" context = await self.get_patient_context(patient_id) prompt = f"""Context: {context} Patient Query: {query} Response:""" async with httpx.AsyncClient() as client: response = await client.post( f"{self.ollama_url}/api/generate", json={"model": "healthcare-agent", "prompt": prompt} ) return response.json()["response"] ```