# ============================================================
# IBM i MCP Client - Environment Configuration
# ============================================================
# Copy this file to .env and configure based on your LLM provider
# cp .env.example .env
# ------------------------------------------------------------
# Option 1: OpenAI (Recommended for Getting Started)
# ------------------------------------------------------------
# Get your API key from: https://platform.openai.com/api-keys
# Cost: Pay per use (see OpenAI pricing)
# Models: gpt-4o, gpt-4-turbo, gpt-3.5-turbo
# OPENAI_API_KEY=sk-proj-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
# ------------------------------------------------------------
# Option 2: IBM WatsonX (Enterprise)
# ------------------------------------------------------------
# Setup:
# 1. Create IBM Cloud account: https://cloud.ibm.com/
# 2. Create WatsonX.ai service instance
# 3. Get API key from: https://cloud.ibm.com/iam/apikeys
# 4. Get project ID from your WatsonX project settings
#
# Models: meta-llama/llama-3-3-70b-instruct, ibm/granite-13b-chat-v2
# IBM_WATSONX_API_KEY=your_watsonx_api_key
# IBM_WATSONX_PROJECT_ID=your_project_id
# IBM_WATSONX_BASE_URL=https://us-south.ml.cloud.ibm.com
# ------------------------------------------------------------
# Option 3: Ollama (Local/Free)
# ------------------------------------------------------------
# Setup:
# 1. Install Ollama: https://ollama.com/download
# 2. Pull a model: ollama pull qwen2.5:latest
# 3. Start Ollama: ollama serve
#
# No API key needed - runs completely locally
# Models: qwen2.5:latest, llama3.2:latest, mistral:latest
# ------------------------------------------------------------
# Option 4: Anthropic Claude (Alternative)
# ------------------------------------------------------------
# Get your API key from: https://console.anthropic.com/
# Cost: Pay per use (see Anthropic pricing)
# Models: claude-3-5-sonnet-20241022, claude-3-opus-20240229
# ANTHROPIC_API_KEY=sk-ant-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
# ============================================================
# Development Settings (Optional)
# ============================================================
# Enable debug logging
# DEBUG=1
# Python log level
# LOG_LEVEL=INFO
# ============================================================
# Usage Examples
# ============================================================
#
# After configuring your .env file, run scripts like:
#
# Basic MCP client (no LLM required):
# uv run mcp_client.py
# uv run list_tool_annotations.py
#
# AI Agent with OpenAI:
# uv run agent.py --model-id "openai:gpt-4o" -p "Show system CPU usage"
#
# AI Agent with WatsonX:
# uv run agent.py --model-id "watsonx:meta-llama/llama-3-3-70b-instruct" -p "List active jobs"
#
# AI Agent with Ollama (local):
# uv run agent.py --model-id "ollama:qwen2.5:latest" -p "Check system status"
#
# Authenticated agent:
# uv run test_auth_agent.py -p "Analyze system performance"
#
# ============================================================
# Security Notes
# ============================================================
# - NEVER commit .env files to version control
# - Add .env to .gitignore
# - Use environment-specific tokens (dev/staging/prod)
# - Rotate API keys and tokens regularly
# - Use short-lived tokens for production environments