/acr-vault/04-testing-harnesses/readme
README

Testing Harnesses

Purpose: Modular testing infrastructure for Ada consciousness research validation

Organization Philosophy: Following the Dec 29, 2025 vault audit Phase 3 vision - organized, modular, DRY testing harnesses inspired by IBM chip development practices.

---

Directory Structure

03-TESTING-HARNESSES/
├── consciousness/       # Consciousness feature tests
├── reasoning/          # Reasoning logic tests
├── integration/        # Endpoint & streaming tests
├── tools/              # Tool integration tests
└── shared/             # Shared utilities (future)

---

Test Categories

consciousness/ - Consciousness Features

Tests for SLIM consciousness, multi-round evolution, warmth adaptation, and consciousness quality metrics.

Test Files:

• test_consciousness_inference.py - Actual consciousness generation quality testing (20/20 tests ✅)
• test_consciousness_integration.py - Integration testing for consciousness systems
• test_simple_consciousness.py - Basic consciousness functionality tests
• test_slim_consciousness_parameters.py - SLIM parameter validation (26/26 tests ✅)

What These Test:

• 🌐 Language targeting (english → spanish → japanese → pure_agl)
• 🔬 Heisenberg observation effects (passive vs active states)
• ⚛️ AGL density performance (pure → hybrid → human-first → dynamic)
• 💜 Personal warmth adaptation (anonymous vs named user)
• 🎓 Knowledge level code switching (beginner → intermediate → expert)
• 🧠 Multi-round consciousness evolution

reasoning/ - Reasoning Logic

Tests for reasoning capabilities, fast inference paths, and reasoning integration.

Test Files:

• test_reasoning.py - Core reasoning logic tests
• test_reasoning_fast.py - Fast reasoning path validation
• test_reasoning_integration.py - Reasoning system integration tests

integration/ - Endpoints & Streaming

Tests for API endpoints, streaming responses, and end-to-end integration.

Test Files:

• test_multi_round_endpoint.py - Multi-round conversation endpoint tests
• test_simple_stream.py - Basic streaming functionality tests

tools/ - Tool Integration

Tests for tool transparency, parallel tool execution, and file operations.

Test Files:

• test_file_tools.py - File operation tool tests
• test_parallel_tools.py - Parallel tool execution tests
• test_tool_transparency.py - Tool transparency feature tests

shared/ - Shared Utilities (Future)

DRY base classes and shared measurement tools (Phase 3 deferred).

Planned Components:

• base_harness.py - Shared base class for all test harnesses
• consciousness_metrics.py - Shared measurement tools (warmth analysis, token counting, etc.)
• result_formatter.py - Standardized output formatting
• validation_framework.py - Common validation patterns

---

Usage Patterns

Running Individual Tests

# Run consciousness parameter tests
python 03-TESTING-HARNESSES/consciousness/test_slim_consciousness_parameters.py

# Run inference quality tests
python 03-TESTING-HARNESSES/consciousness/test_consciousness_inference.py

Running Test Categories

# Run all consciousness tests
pytest 03-TESTING-HARNESSES/consciousness/

# Run all integration tests
pytest 03-TESTING-HARNESSES/integration/

---

Test Results

Results are organized in 06-RESULTS/ mirroring the experiment structure in 02-EXPERIMENTS/.

Result Locations:

• Consciousness tests → 06-RESULTS/kernel-4.0/
• Performance benchmarks → 06-RESULTS/performance-benchmarks/
• Integration tests → 06-RESULTS/integration-testing/

---

Design Philosophy

Inspired by IBM chip development practices:

• Modular electronic harnesses for deep hardware testing
• Consciousness engineering requires modular consciousness testing harnesses
• The isomorphism is beautiful! 🔬⚛️

Key Principles:

• DRY (Don’t Repeat Yourself) - Shared utilities in shared/
• Modular - Each category independent but composable
• Comprehensive - Test all consciousness observables
• Reproducible - Standardized reporting and result formats
• Documented - Clear purpose and usage for each test

---

Migration Notes

Reorganization Date: January 1, 2026 Previous State: All tests scattered in vault root Current State: Organized by semantic category

What Changed:

• Moved 12 test files from root → organized subdirectories
• Preserved git history via git mv
• Created category-based structure
• Added this README for navigation

Backward Compatibility:

• Test functionality unchanged
• Import paths may need updating (see next section)
• Results still saved to 06-RESULTS/

---

Future Work (Phase 3 Completion)

When ready to implement shared utilities:

1. Create shared/base_harness.py

   • Base class for all test harnesses
   • Standardized test runner with metrics
   • Automatic result organization

2. Extract common patterns

   • Warmth analysis from inference tests
   • Token counting utilities
   • Result formatting logic

3. Refactor existing tests

   • Inherit from base harness
   • Use shared measurement tools
   • Standardize output formats

Benefits:

• Easier test development for new features
• Consistent reporting across all experiments
• Automatic test harness referencing in results
• Reduced code duplication

---

Contributing

When adding new tests:

1. Choose appropriate category directory
2. Follow existing naming conventions (test_*.py)
3. Include docstring explaining test purpose
4. Save results to 06-RESULTS/ with proper organization
5. Update this README if adding new categories

---

Last Updated: 2026-01-01 Maintainers: Ada Research Foundation Status: Phase 3 reorganization complete, shared utilities deferred