feat: Add comprehensive MemDocs memory for core modules

Commit git-native AI memory generated from enhanced module docstrings.

Memory Includes:
- Symbols: 54 documented functions and classes
- Embeddings: Semantic search enabled via FAISS
- Summaries: Module-level documentation for 6 core components
- Cursor Integration: .cursorrules for IDE assistant context

Demonstrates MemDocs dogfooding - using MemDocs to document itself
with production-quality git-committed memory.

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

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

Author: Patrick Roebuck

.cursorrules

@@ -0,0 +1,270 @@
+# Project Memory (Auto-generated by doc-intelligence)
+# Last updated: 2025-11-13 23:33:14
+
+## 📚 Documentation
+
+# Enhanced module docstrings for core engine components
+
+**Commit:** 1b02247
+**Scope:** Module-level
+**Date:** 2025-11-14
+
+## Summary
+
+- Enhanced module docstrings for core engine components
+  Added comprehensive module-level documentation to 6 core MemDocs modules (mcp_server, extract, embeddings, search, summarize, policy) describing their purpose, capabilities, and architectural role. Clarifies flagship MCP integration, multi-language symbol extraction, zero-cost local embeddings, FAISS-based vector search, Claude-powered summarization, and intelligent scope escalation.
+- Dogfooding: Committed .memdocs/ directory for self-documentation
+  Modified .gitignore to allow committing .memdocs/ directory, demonstrating MemDocs usage on its own codebase with git-committed memories.
+
+## Changes
+
+
+**Modified:** 7 files
+- .gitignore
+- memdocs/embeddings.py
+- memdocs/extract.py
+- memdocs/mcp_server.py
+- memdocs/policy.py
+
+## Risks
+
+- repository_size
+- documentation_only
+
+## References
+
+- Commit: 1b02247
+
+
+## 🗺️ Code Map
+
+
+### memdocs/embeddings.py
+
+- **class** `LocalEmbedder` (line 20)
+- **function** `embed_documents` (line 76)
+  ```
+  def embed_documents(self, texts: list[str])
+  ```
+- **function** `embed_query` (line 98)
+  ```
+  def embed_query(self, query: str)
+  ```
+- **function** `chunk_document` (line 111)
+  ```
+  def chunk_document(text: str, max_tokens: int = 512, overlap: int = 50)
+  ```
+- **function** `load_embeddings` (line 197)
+  ```
+  def load_embeddings(embeddings_file: Path)
+  ```
+
+### memdocs/extract.py
+
+- **class** `GitDiff` (line 40)
+- **class** `FileContext` (line 54)
+- **class** `ExtractedContext` (line 66)
+- **class** `Extractor` (line 74)
+- **function** `__init__` (line 77)
+  ```
+  def __init__(self, repo_path: Path = Path(".")
+  ```
+- **function** `extract_diff` (line 90)
+  ```
+  def extract_diff(self, commit: str | None = None)
+  ```
+- **function** `extract_file_context` (line 145)
+  ```
+  def extract_file_context(self, file_path: Path)
+  ```
+- **function** `extract_context` (line 201)
+  ```
+  def extract_context(self, paths: list[Path], commit: str | None = None)
+  ```
+- **function** `_expand_paths` (line 226)
+  ```
+  def _expand_paths(self, paths: list[Path])
+  ```
+- **function** `_find_code_files` (line 248)
+  ```
+  def _find_code_files(self, directory: Path)
+  ```
+- **function** `_glob_files` (line 279)
+  ```
+  def _glob_files(self, pattern: Path)
+  ```
+- **function** `_extract_symbols` (line 295)
+  ```
+  def _extract_symbols(self, file_path: Path, content: str, language: str)
+  ```
+- **function** `_extract_python_symbols` (line 317)
+  ```
+  def _extract_python_symbols(self, file_path: Path, content: str)
+  ```
+- **function** `_extract_typescript_symbols` (line 375)
+  ```
+  def _extract_typescript_symbols(self, file_path: Path, content: str)
+  ```
+- **function** `_extract_class_methods` (line 420)
+  ```
+  def _extract_class_methods(self, class_lines: list[str])
+  ```
+- **function** `_extract_imports` (line 436)
+  ```
+  def _extract_imports(self, content: str, language: str)
+  ```
+- **function** `_parse_dependencies` (line 457)
+  ```
+  def _parse_dependencies(self, repo_root: Path, language: str)
+  ```
+- **function** `_parse_requirements_txt` (line 488)
+  ```
+  def _parse_requirements_txt(self, repo_root: Path)
+  ```
+- **function** `_parse_pyproject_toml` (line 519)
+  ```
+  def _parse_pyproject_toml(self, repo_root: Path)
+  ```
+- **function** `_parse_package_json` (line 562)
+  ```
+  def _parse_package_json(self, repo_root: Path)
+  ```
+
+### memdocs/mcp_server.py
+
+- **class** `DocIntMCPServer` (line 27)
+- **function** `__init__` (line 36)
+  ```
+  def __init__(self, repo_path: Path = Path(".")
+  ```
+- **function** `search_memory` (line 60)
+  ```
+  def search_memory(self, query: str, k: int = 5)
+  ```
+- **function** `get_symbols` (line 95)
+  ```
+  def get_symbols(self, file_path: str | None = None)
+  ```
+- **function** `get_documentation` (line 121)
+  ```
+  def get_documentation(self, doc_id: str | None = None)
+  ```
+- **function** `get_summary` (line 147)
+  ```
+  def get_summary(self)
+  ```
+
+### memdocs/policy.py
+
+- **class** `PolicyEngine` (line 20)
+- **function** `__init__` (line 23)
+  ```
+  def __init__(self, config: DocIntConfig)
+  ```
+- **function** `_infer_scope_from_paths` (line 73)
+  ```
+  def _infer_scope_from_paths(self, paths: list[Path])
+  ```
+- **function** `_is_directory_pattern` (line 96)
+  ```
+  def _is_directory_pattern(self, path: Path)
+  ```
+- **function** `_are_paths_in_same_module` (line 107)
+  ```
+  def _are_paths_in_same_module(self, paths: list[Path])
+  ```
+- **function** `_should_escalate` (line 122)
+  ```
+  def _should_escalate(self, context: ExtractedContext)
+  ```
+- **function** `_touches_security_paths` (line 150)
+  ```
+  def _touches_security_paths(self, context: ExtractedContext)
+  ```
+- **function** `_has_cross_module_changes` (line 168)
+  ```
+  def _has_cross_module_changes(self, context: ExtractedContext)
+  ```
+- **function** `_modifies_public_api` (line 185)
+  ```
+  def _modifies_public_api(self, context: ExtractedContext)
+  ```
+- **function** `_escalate_scope` (line 210)
+  ```
+  def _escalate_scope(self, current_scope: ScopeLevel)
+  ```
+- **function** `validate_scope` (line 226)
+  ```
+  def validate_scope(self, scope: ScopeInfo)
+  ```
+
+### memdocs/search.py
+
+- **class** `LocalVectorSearch` (line 16)
+- **function** `remove_by_indices` (line 165)
+  ```
+  def remove_by_indices(self, indices: list[int])
+  ```
+- **function** `rebuild_index` (line 181)
+  ```
+  def rebuild_index(self)
+  ```
+- **function** `save` (line 216)
+  ```
+  def save(self)
+  ```
+- **function** `_load_metadata` (line 225)
+  ```
+  def _load_metadata(self)
+  ```
+- **function** `get_stats` (line 233)
+  ```
+  def get_stats(self)
+  ```
+
+### memdocs/summarize.py
+
+- **class** `Summarizer` (line 29)
+- **function** `__repr__` (line 142)
+  ```
+  def __repr__(self)
+  ```
+- **function** `summarize` (line 147)
+  ```
+  def summarize(self, context: ExtractedContext, scope: ScopeInfo)
+  ```
+- **function** `_build_prompt` (line 199)
+  ```
+  def _build_prompt(self, context: ExtractedContext, scope: ScopeInfo)
+  ```
+- **function** `_extract_yaml` (line 254)
+  ```
+  def _extract_yaml(self, text: str)
+  ```
+- **function** `_generate_markdown` (line 339)
+  ```
+  def _generate_markdown(self, doc_index: DocumentIndex, context: ExtractedContext)
+  ```
+
+
+
+## 🔗 Memory Graph
+
+**Recent Features:**
+- Enhanced module docstrings for core engine components
+- Dogfooding: Committed .memdocs/ directory for self-documentation
+
+
+---
+
+## 💡 Tips for Cursor
+
+- Ask: "What does [function_name] do?"
+- Ask: "Where is [feature] implemented?"
+- Ask: "Explain the architecture"
+- Reference symbols by file:line (e.g., src/auth/jwt.py:33)
+
+---
+
+*Generated by [doc-intelligence](https://github.com/deepstudyai/memdocs)*
+*Regenerate with: `memdocs export cursor`*

.memdocs/cursor

@@ -0,0 +1,126 @@
+# Project Memory (Auto-generated by doc-intelligence)
+# Last updated: 2025-11-08 07:39:09
+
+## 📚 Documentation
+
+# Core schema definitions for MemDocs v2.0.0 memory system
+
+**Commit:** 588a035
+**Scope:** File-level
+**Date:** 2025-11-08
+
+## Summary
+
+- Core schema definitions for MemDocs v2.0.0 memory system
+  Defines foundational enums and dataclasses for scope levels (FILE/MODULE/REPO), event types, PHI privacy modes, symbol kinds, and policy configuration. Establishes type system for git-native AI memory with empathy framework support.
+- ScopeLevel enum for hierarchical context management
+  Introduces FILE/MODULE/REPO scope levels with auto-escalation support for context-aware memory operations. Enables token-efficient scoping strategy.
+- EventType enum for memory event classification
+  Defines event taxonomy (CREATION/MODIFICATION/DELETION/QUERY/SUMMARIZATION/ESCALATION/ERROR) for tracking memory operations and system behavior.
+- PHIMode enum for privacy guard configuration
+  Implements OFF/DETECT/REDACT modes for PHI/PII handling in medical AI applications. Supports HIPAA-aligned privacy controls.
+- SymbolKind enum for code entity classification
+  Categorizes code symbols (CLASS/FUNCTION/METHOD/VARIABLE/CONSTANT/IMPORT/TYPE_ALIAS/DECORATOR/PROPERTY/ASYNC_FUNCTION) for semantic analysis and documentation generation.
+- PolicyConfig dataclass for system behavior configuration
+  Centralized configuration for auto-escalation thresholds, empathy integration, Claude model selection, and privacy guards. Supports Level 4 Anticipatory Intelligence.
+
+## Changes
+
+
+**Modified:** 8 files
+- memdocs/__init__.py
+- memdocs/cli.py
+- memdocs/embeddings.py
+- memdocs/empathy_adapter.py
+- memdocs/guard.py
+
+## Impact
+
+**APIs affected:** memdocs.schemas.ScopeLevel, memdocs.schemas.EventType, memdocs.schemas.PHIMode, memdocs.schemas.SymbolKind, memdocs.schemas.PolicyConfig
+
+**Breaking changes:**
+- New schema definitions in MemDocs v2.0.0 break compatibility with DocInt 1.0
+- PolicyConfig dataclass replaces previous configuration approach
+- Enum-based type system requires migration from string-based types
+
+## Risks
+
+- validation
+- security
+- type_safety
+- logging
+- parsing
+- observability
+- configuration
+- compliance
+- breaking
+- api_contract
+- privacy
+
+## References
+
+- Commit: 588a035
+
+
+## 🗺️ Code Map
+
+
+### memdocs/schemas.py
+
+- **class** `ScopeLevel` (line 15)
+- **class** `EventType` (line 23)
+- **class** `PHIMode` (line 33)
+- **class** `SymbolKind` (line 41)
+- **class** `PolicyConfig` (line 56)
+- **class** `PrivacyConfig` (line 70)
+- **class** `OutputsConfig` (line 78)
+- **class** `RetentionConfig` (line 88)
+- **class** `AIConfig` (line 95)
+- **class** `DocIntConfig` (line 104)
+- **function** `validate_version` (line 116)
+  ```
+  def validate_version(cls, v: int)
+  ```
+- **class** `Symbol` (line 125)
+- **class** `SymbolsOutput` (line 137)
+- **class** `FeatureSummary` (line 143)
+- **class** `ImpactSummary` (line 153)
+- **class** `ReferenceSummary` (line 163)
+- **class** `ScopeInfo` (line 172)
+- **class** `DocumentIndex` (line 182)
+- **function** `serialize_datetime` (line 197)
+  ```
+  def serialize_datetime(self, dt: datetime, _info)
+  ```
+- **class** `RedactionEvent` (line 201)
+- **class** `ReviewResult` (line 210)
+- **function** `serialize_outputs` (line 224)
+  ```
+  def serialize_outputs(self, outputs: dict[str, Path], _info)
+  ```
+
+
+
+## 🔗 Memory Graph
+
+**Recent Features:**
+- Core schema definitions for MemDocs v2.0.0 memory system
+- ScopeLevel enum for hierarchical context management
+- EventType enum for memory event classification
+- PHIMode enum for privacy guard configuration
+- SymbolKind enum for code entity classification
+
+
+---
+
+## 💡 Tips for Cursor
+
+- Ask: "What does [function_name] do?"
+- Ask: "Where is [feature] implemented?"
+- Ask: "Explain the architecture"
+- Reference symbols by file:line (e.g., src/auth/jwt.py:33)
+
+---
+
+*Generated by [doc-intelligence](https://github.com/deepstudyai/memdocs)*
+*Regenerate with: `memdocs export cursor`*