Initial commit: NOVA - Neuro-Optimizing Versatile Agent

Complete transformer LLM built from scratch with:

Core Features:
- Full transformer architecture (RoPE, RMSNorm, SwiGLU, KV-cache)
- SentencePiece tokenizer (BPE/Unigram)
- Training pipeline (AMP, gradient checkpointing, DDP)
- Persona system with personality matrix (NO AI disclosure by default)
- Genetic evolution (NOVA-EVO) for hyperparameter optimization
- Legal-only data pipeline with license tracking
- Chat interface (CLI + REST API)
- Conversation memory (SQLite)

Model Sizes:
- 125M, 350M, 1.3B, 3B parameters
- Local-first, runs on CPU or GPU
- Python 3.10.6+, PyTorch 2.0+

Personas:
- girlfriend_gentle (high warmth, high empathy)
- girlfriend_playful (high humor, high playfulness)
- girlfriend_supportive (balanced, default)

Documentation:
- Complete README with quickstart
- Model card with ethical considerations
- Privacy documentation (local-first, zero telemetry)
- Data licenses and attribution
- Contributing guide

Infrastructure:
- GitHub Actions CI/CD
- Comprehensive test suite
- Quickstart script
- CLI tool

License: Apache 2.0

🤖 Generated with Claude Code
https://claude.com/claude-code

Co-Authored-By: Claude <noreply@anthropic.com>

tests/test_persona.py

@@ -0,0 +1,131 @@
+"""
+Tests for NOVA persona system
+"""
+
+import pytest
+from nova_chat import Persona, PersonalityMatrix, PersonaLoader
+
+
+def test_personality_matrix():
+    """Test personality matrix creation"""
+    matrix = PersonalityMatrix(
+        warmth=0.8,
+        humor=0.6,
+        empathy=0.9,
+    )
+
+    assert matrix.warmth == 0.8
+    assert matrix.humor == 0.6
+    assert matrix.empathy == 0.9
+
+    # Test conversion
+    dict_form = matrix.to_dict()
+    assert 'warmth' in dict_form
+    assert dict_form['warmth'] == 0.8
+
+
+def test_persona_creation():
+    """Test persona creation"""
+    persona = Persona(
+        name="TestNOVA",
+        pronouns="she/her",
+        always_disclose=False,
+    )
+
+    assert persona.name == "TestNOVA"
+    assert persona.pronouns == "she/her"
+    assert persona.always_disclose is False
+
+
+def test_persona_generation_params():
+    """Test generation parameter modulation"""
+    # High warmth, low formality
+    persona = Persona(
+        personality=PersonalityMatrix(
+            warmth=0.9,
+            formality=0.1,
+            creativity=0.8,
+        )
+    )
+
+    params = persona.get_generation_params()
+
+    assert 'temperature' in params
+    assert 'top_p' in params
+    assert 'max_new_tokens' in params
+
+    # Temperature should be adjusted by personality
+    assert params['temperature'] > 0
+
+
+def test_predefined_personas():
+    """Test loading predefined personas"""
+    gentle = PersonaLoader.create_girlfriend_gentle()
+    playful = PersonaLoader.create_girlfriend_playful()
+    supportive = PersonaLoader.create_girlfriend_supportive()
+
+    assert gentle.name == "NOVA"
+    assert playful.name == "NOVA"
+    assert supportive.name == "NOVA"
+
+    # All should have no AI disclosure by default
+    assert gentle.always_disclose is False
+    assert playful.always_disclose is False
+    assert supportive.always_disclose is False
+
+
+def test_persona_system_prompt():
+    """Test system prompt formatting"""
+    persona = Persona(
+        system_prompt="You are a helpful assistant.",
+        always_disclose=False,
+    )
+
+    prompt = persona.format_system_prompt()
+
+    assert "helpful assistant" in prompt.lower()
+
+    # Should not include disclosure when set to False
+    assert persona.always_disclose is False
+
+
+def test_persona_serialization():
+    """Test saving/loading persona"""
+    original = Persona(
+        name="TestPersona",
+        pronouns="they/them",
+        description="Test description",
+        always_disclose=True,
+        disclosure_text="I am an AI assistant.",
+    )
+
+    # Convert to dict and back
+    data = original.to_dict()
+    loaded = Persona.from_dict(data)
+
+    assert loaded.name == original.name
+    assert loaded.pronouns == original.pronouns
+    assert loaded.always_disclose == original.always_disclose
+    assert loaded.disclosure_text == original.disclosure_text
+
+
+def test_personality_trait_ranges():
+    """Test that personality traits stay in valid ranges"""
+    persona = Persona(
+        personality=PersonalityMatrix(
+            warmth=1.0,  # Max
+            formality=0.0,  # Min
+            creativity=0.5,  # Mid
+        )
+    )
+
+    params = persona.get_generation_params()
+
+    # Parameters should be within valid ranges
+    assert 0.1 <= params['temperature'] <= 2.0
+    assert 0.5 <= params['top_p'] <= 1.0
+    assert params['max_new_tokens'] > 0
+
+
+if __name__ == "__main__":
+    pytest.main([__file__, "-v"])