tests/README.md

Tests

This directory contains all tests for the LLM Model Therapist Tool.

Structure

tests/
├── test_evaluators/        # Evaluator tests
│   ├── test_talk_type_evaluator.py
│   └── ...
├── test_parsers/           # Parser tests
│   └── test_conversation_parser.py
├── test_services/          # Service layer tests
│   ├── test_orchestrator.py
│   └── test_key_manager.py
├── test_providers/         # Provider client tests
│   ├── test_openai_client.py
│   └── test_huggingface_client.py
└── fixtures/               # Test data
    ├── sample_conversations/
    └── mock_responses/

Running Tests

Run all tests

cd /Users/ryan/Dev/LLM_Model_Therapist_Tool/web
python -m pytest tests/

Run specific test module

python -m pytest tests/test_evaluators/test_talk_type_evaluator.py

Run a specific test file directly

cd /Users/ryan/Dev/LLM_Model_Therapist_Tool/web
python tests/test_evaluators/test_talk_type_evaluator.py

Test with command-line arguments

# Test single utterance
python tests/test_evaluators/test_talk_type_evaluator.py "I want to quit smoking"

# Test with debug mode
python tests/test_evaluators/test_talk_type_evaluator.py --debug "I want to quit smoking"

Test Conventions

• Test files should be named test_<module_name>.py
• Test functions should be named test_<functionality>()
• Use fixtures from fixtures/ directory for sample data
• Mock external API calls when possible to avoid rate limits
• Use environment variables for API keys (fallback to test keys for CI)

Environment Variables

export HF_API_KEY="your_huggingface_key"
export OPENAI_API_KEY="your_openai_key"