SOP-001: Environment Setup

Document Control

Property	Value
SOP ID	001
Title	Environment Setup
Version	1.0
Status	Active
Complexity	Easy

Purpose

Install all required dependencies and configure the development environment for building self-evolving agents.

Prerequisites

• Python 3.10 or higher
• pip package manager
• Terminal/command line access
• OpenAI account with API access

Procedure Flowchart

┌─────────────────────────────────────────────────────────────────────┐
│                    ENVIRONMENT SETUP FLOW                            │
├─────────────────────────────────────────────────────────────────────┤
│                                                                      │
│  ┌─────────────────┐                                                │
│  │  Check Python   │                                                │
│  │    Version      │                                                │
│  └────────┬────────┘                                                │
│           │                                                          │
│           ▼                                                          │
│  ┌─────────────────┐     ┌─────────────────┐                        │
│  │ Install Core    │────►│ Install Optional│                        │
│  │  Dependencies   │     │  Dependencies   │                        │
│  └────────┬────────┘     └────────┬────────┘                        │
│           │                       │                                  │
│           └───────────┬───────────┘                                  │
│                       ▼                                              │
│           ┌─────────────────┐                                       │
│           │  Create .env    │                                       │
│           │     File        │                                       │
│           └────────┬────────┘                                       │
│                    │                                                 │
│                    ▼                                                 │
│           ┌─────────────────┐                                       │
│           │    Verify       │                                       │
│           │  Installation   │                                       │
│           └─────────────────┘                                       │
│                                                                      │
└─────────────────────────────────────────────────────────────────────┘

Step-by-Step Procedure

Step 1: Verify Python Version

python --version
# Expected: Python 3.10.x or higher

WARNING

Python 3.9 and below are not supported due to type hint syntax requirements.

Step 2: Create Project Directory

mkdir self-evolving-agents
cd self-evolving-agents

Step 3: Create Virtual Environment (Recommended)

python -m venv venv

# Windows
venv\Scripts\activate

# macOS/Linux
source venv/bin/activate

Step 4: Install Core Dependencies

pip install --upgrade openai openai-agents pydantic pandas python-dotenv

Package Descriptions:

Package	Version	Purpose
openai	Latest	OpenAI API client
openai-agents	Latest	Agents SDK for building agents
pydantic	Latest	Data validation for PromptVersionEntry
pandas	Latest	Dataset handling
python-dotenv	Latest	Environment variable management

Step 5: Install Optional Dependencies (GEPA)

pip install gepa litellm

Package	Purpose
gepa	Genetic-Pareto prompt optimization
litellm	Multi-model support for GEPA

Step 6: Configure Environment Variables

Create .env file in project root:

# .env
OPENAI_API_KEY=sk-your-api-key-here

Step 7: Create Configuration File

Create config.py:

import os
from dotenv import load_dotenv

load_dotenv()

# API Configuration
OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")

# Model Configuration
DEFAULT_MODEL = "gpt-5"
EVAL_MODEL = "gpt-4.1"

# Optimization Thresholds
LENIENT_PASS_RATIO = 0.75
LENIENT_AVERAGE_THRESHOLD = 0.85
MAX_OPTIMIZATION_RETRIES = 3

Step 8: Verify Installation

Create verify_setup.py:

import os
from dotenv import load_dotenv
from openai import OpenAI

load_dotenv()

def verify():
    # Check API key
    api_key = os.getenv("OPENAI_API_KEY")
    if not api_key:
        print("❌ OPENAI_API_KEY not found in .env")
        return False

    # Test API connection
    try:
        client = OpenAI(api_key=api_key)
        models = client.models.list()
        print("✅ OpenAI API connection successful")
        print(f"✅ Found {len(list(models))} available models")
        return True
    except Exception as e:
        print(f"❌ API connection failed: {e}")
        return False

if __name__ == "__main__":
    verify()

Run verification:

python verify_setup.py

Expected Output:

✅ OpenAI API connection successful
✅ Found X available models

Verification Checklist

• [ ] Python 3.10+ installed and verified
• [ ] Virtual environment created and activated
• [ ] Core dependencies installed (openai, openai-agents, pydantic, pandas, python-dotenv)
• [ ] Optional dependencies installed (gepa, litellm) if using GEPA
• [ ] .env file created with valid API key
• [ ] Verification script runs successfully

Troubleshooting

Issue: ModuleNotFoundError: No module named 'openai'

Solution: Ensure virtual environment is activated and reinstall:

pip install --force-reinstall openai

Issue: API key not found

Solution: Verify .env file location and format:

# Check file exists
ls -la .env

# Check content (careful not to expose key)
head -c 20 .env

Issue: SSL Certificate errors

Solution: Update certificates:

pip install --upgrade certifi

See Also

• SOP-002: Dataset Preparation
• SOP-003: Eval Creation
• Quick Reference: Commands