test: Add extended tests for PII scrubber and Claude memory

- test_pii_scrubber_extended.py: 86 new tests for custom patterns,
  overlapping PII detection, healthcare variants, and edge cases
- test_claude_memory_extended.py: 43 new tests for enterprise/user
  memory loading, error handling, and import processing
- TEST_COVERAGE_PLAN.md: Coverage improvement roadmap

Total tests: 1,491 → 1,620 (+129)
Coverage: 75.90% → 77.88% (+2%)

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <[email protected]>

Author: 2 people

TEST_COVERAGE_PLAN.md

@@ -0,0 +1,169 @@
+# Test Coverage Improvement Plan
+
+**Current:** 75.90% (1,491 tests)
+**Target:** 90%+
+**Estimated New Tests:** ~150-180 test cases
+
+---
+
+## Phase 1: Security Modules (Priority: CRITICAL)
+
+### 1.1 PII Scrubber (60% → 90%)
+**Estimated Tests:** 25
+
+| Test Category | Tests | Parallelizable |
+|---------------|-------|----------------|
+| Custom pattern management | 8 | Yes |
+| Overlapping PII detection | 5 | Yes |
+| Pattern validation suite | 8 | Yes |
+| Healthcare PII variants | 4 | Yes |
+
+### 1.2 Secure MemDocs (75% → 95%)
+**Estimated Tests:** 30
+
+| Test Category | Tests | Parallelizable |
+|---------------|-------|----------------|
+| Encryption/decryption operations | 10 | Yes |
+| Pattern classification logic | 8 | Yes |
+| Access control enforcement | 8 | Yes |
+| Retention policy validation | 4 | Yes |
+
+### 1.3 Audit Logger (73% → 92%)
+**Estimated Tests:** 25
+
+| Test Category | Tests | Parallelizable |
+|---------------|-------|----------------|
+| Directory initialization | 4 | Yes |
+| Log rotation and cleanup | 8 | Yes |
+| Query operators | 8 | Yes |
+| Compliance reporting | 5 | Yes |
+
+---
+
+## Phase 2: Claude Memory Integration (72% → 95%)
+
+**Estimated Tests:** 20
+
+| Test Category | Tests | Parallelizable |
+|---------------|-------|----------------|
+| Enterprise memory loading | 5 | Yes |
+| User memory loading | 4 | Yes |
+| Error handling paths | 6 | Yes |
+| Import processing edge cases | 5 | Yes |
+
+---
+
+## Phase 3: CLI Modules
+
+### 3.1 empathy_software_plugin/cli.py (34% → 75%)
+**Estimated Tests:** 35
+
+| Test Category | Tests | Parallelizable |
+|---------------|-------|----------------|
+| analyze_project() async | 10 | No (integration) |
+| gather_project_context() | 8 | Yes |
+| scan_command() | 10 | Yes |
+| list_wizards/wizard_info | 7 | Yes |
+
+### 3.2 src/empathy_os/cli.py (45% → 80%)
+**Estimated Tests:** 40
+
+| Test Category | Tests | Parallelizable |
+|---------------|-------|----------------|
+| cmd_run() interactive REPL | 12 | No (sequential) |
+| cmd_inspect() branches | 8 | Yes |
+| cmd_export/import | 10 | Yes |
+| cmd_wizard() flow | 10 | Yes |
+
+---
+
+## Parallel Execution Strategy
+
+### Batch 1 (Security - Can Run Simultaneously)
+```
+Agent A: PII Scrubber tests (25 tests)
+Agent B: Secure MemDocs tests (30 tests)
+Agent C: Audit Logger tests (25 tests)
+Agent D: Claude Memory tests (20 tests)
+```
+
+### Batch 2 (CLI - Sequential Dependencies)
+```
+Agent E: Software plugin CLI (35 tests)
+Agent F: EmpathyOS CLI (40 tests)
+```
+
+---
+
+## Test Fixtures Required
+
+### Shared Fixtures (Create First)
+```python
+# conftest.py additions
+
+@pytest.fixture
+def temp_project_with_ai_files(tmp_path):
+    """Project with AI library imports"""
+
+@pytest.fixture
+def mock_registry():
+    """Mock plugin registry"""
+
+@pytest.fixture
+def mock_empathy_os():
+    """Mock EmpathyOS instance"""
+
+@pytest.fixture
+def simulate_user_input(monkeypatch):
+    """Interactive input simulation"""
+
+@pytest.fixture
+def encryption_key():
+    """Valid AES-256 key for testing"""
+
+@pytest.fixture
+def mock_audit_directory(tmp_path):
+    """Temporary audit log directory"""
+```
+
+---
+
+## Execution Order
+
+1. **Create fixtures** (conftest.py) - 30 min
+2. **Phase 1 Batch** (Security) - Parallel agents - 2 hours
+3. **Phase 2** (Claude Memory) - Single agent - 1 hour
+4. **Phase 3 Batch** (CLI) - Parallel agents - 2 hours
+5. **Integration tests** - Sequential - 1 hour
+6. **Coverage verification** - Final run - 30 min
+
+**Total estimated time: 6-7 hours**
+
+---
+
+## Success Criteria
+
+| Module | Before | After | Delta |
+|--------|--------|-------|-------|
+| pii_scrubber.py | 60% | 90% | +30% |
+| secure_memdocs.py | 75% | 95% | +20% |
+| audit_logger.py | 73% | 92% | +19% |
+| claude_memory.py | 72% | 95% | +23% |
+| software_plugin/cli.py | 34% | 75% | +41% |
+| empathy_os/cli.py | 45% | 80% | +35% |
+| **OVERALL** | **75.9%** | **90%+** | **+14%** |
+
+---
+
+## Agent Task Assignments
+
+When ready to execute, spawn these agents in parallel:
+
+```
+Task 1: "Write tests for PII scrubber custom patterns and overlaps"
+Task 2: "Write tests for secure_memdocs encryption and access control"
+Task 3: "Write tests for audit_logger rotation and compliance"
+Task 4: "Write tests for claude_memory enterprise/user loading"
+Task 5: "Write tests for software_plugin CLI commands"
+Task 6: "Write tests for empathy_os CLI interactive commands"
+```