Test Coverage Audit¶

Audit Date: 2026-01-10 Total Test Files: 44 Total Test Functions: 726

Executive Summary¶

The Null Terminal test suite has good foundational coverage but has critical gaps in key areas. The codebase follows testing best practices with proper fixtures and async patterns.

Strengths¶

• Comprehensive unit tests for config, AI factory, and storage
• Good integration test coverage for UI interactions
• Proper use of mock_home fixture to protect user data
• Async testing patterns with pytest-asyncio

Critical Gaps¶

• No tests for ai/ provider implementations (only factory tested)
• Minimal handlers/ai_executor.py testing (complexity hotspot)
• No tests for most widgets (blocks, sidebar, status_bar)
• No tests for screens (provider, config, prompts screens)

Coverage by Module¶

Well-Covered (>70%)¶

Moderately Covered (30-70%)¶

Poorly Covered (<30%)¶

No Test Coverage¶

Integration Test Coverage¶

Good Coverage¶

• App mounting and structure (test_app.py)
• Block creation and rendering (test_blocks.py)
• Input handling and modes (test_input.py)
• Keybindings (test_keybindings.py)
• Screen opening/closing (test_screens.py)
• Theme switching (test_themes.py)

Missing Integration Tests¶

• AI response streaming and rendering
• Agent mode multi-step execution
• MCP tool discovery and execution
• Session save/load/export
• Provider switching
• Context window management
• File tree sidebar interaction

Critical Test Gaps¶

1. AI Executor (handlers/ai_executor.py)¶

Current: 3 tests Needed: 20+ tests

# Missing test scenarios:
- test_streaming_text_response
- test_streaming_with_thinking_extraction
- test_tool_call_approval_flow
- test_tool_call_rejection_flow
- test_agent_mode_iteration
- test_agent_mode_max_iterations
- test_agent_mode_cancellation
- test_context_building
- test_error_handling_during_stream
- test_token_usage_tracking
- test_cost_calculation

2. Provider Implementations (ai/*.py)¶

Current: 0 tests per provider Needed: Per-provider test files

# Each provider should test:
- test_validate_connection_success
- test_validate_connection_failure
- test_list_models
- test_generate_streaming
- test_generate_with_tools
- test_error_handling
- test_api_key_validation

3. Block Widgets (widgets/blocks/*.py)¶

Current: Integration only Needed: Unit tests for widget logic

# Missing widget tests:
- test_ai_response_block_rendering
- test_thinking_widget_toggle
- test_tool_accordion_expand_collapse
- test_iteration_widget_nested_tools
- test_command_block_exit_code_display

4. Context Manager (context.py)¶

Current: 0 tests Needed: Core functionality tests

# Missing context tests:
- test_build_messages_from_blocks
- test_context_truncation
- test_token_estimation
- test_message_ordering

Recommendations¶

Immediate Priority (Week 1)¶

1. Add context.py tests - Core functionality with no coverage
2. Expand ai_executor tests - Most complex code path
3. Add provider mock tests - Test with mocked HTTP responses

Short Term (Month 1)¶

1. Widget unit tests - Test block widget logic
2. Screen tests - Test modal screen interactions
3. Command tests - Test slash command handlers

Long Term¶

1. End-to-end tests - Full user flow testing
2. Performance tests - Streaming latency, UI responsiveness
3. Snapshot tests - UI regression detection

Test Infrastructure Notes¶

Existing Fixtures¶

Missing Fixtures Needed¶

# Proposed new fixtures:

@pytest.fixture
def mock_provider():
    """Create a mock LLMProvider with streaming support."""
    pass

@pytest.fixture
def mock_streaming_response():
    """Generate mock StreamChunk sequence."""
    pass

@pytest.fixture
async def app_with_ai_response(running_app):
    """App with a completed AI response block."""
    pass

Test Organization¶

Current structure is good. Suggested additions:

tests/
├── unit/
│   ├── widgets/          # NEW: Widget unit tests
│   │   ├── test_blocks.py
│   │   ├── test_sidebar.py
│   │   └── test_status_bar.py
│   └── ai/
│       ├── test_ollama.py    # NEW: Provider tests
│       └── test_anthropic.py
├── integration/
│   ├── test_ai_flow.py       # NEW: AI interaction tests
│   └── test_agent_flow.py    # NEW: Agent mode tests
└── e2e/                      # NEW: End-to-end tests
    └── test_user_flows.py

Coverage Goals¶

Running Coverage¶

# Generate coverage report
uv run pytest --cov=. --cov-report=html --cov-report=term-missing

# View HTML report
open htmlcov/index.html

# Coverage for specific module
uv run pytest --cov=handlers --cov-report=term-missing tests/unit/handlers/