mcp-memory-service

# MCP Memory Service [](https://opensource.org/licenses/Apache-2.0) [](https://pypi.org/project/mcp-memory-service/) [](https://pypi.org/project/mcp-memory-service/) [](https://github.com/doobidoo/mcp-memory-service/actions/workflows/main.yml) [](https://pypi.org/project/mcp-memory-service/) [](https://github.com/doobidoo/mcp-memory-service/commits/main) [](https://github.com/doobidoo/mcp-memory-service/stargazers) [](https://github.com/doobidoo/mcp-memory-service#-in-production) [](https://claude.ai) [](https://cursor.sh) [](https://modelcontextprotocol.io/) [](https://github.com/doobidoo/mcp-memory-service/wiki) [](https://github.com/doobidoo/mcp-memory-service/pkgs/container/mcp-memory-service) [](https://github.com/doobidoo/mcp-memory-service/issues) [](https://deepwiki.com/doobidoo/mcp-memory-service) **Production-ready MCP memory service** with **zero database locks**, **hybrid backend** (fast local + cloud sync), and **intelligent memory search** for **AI assistants**. Features **v8.9.0 auto-configuration** for multi-client access, **5ms local reads** with background Cloudflare sync, **Natural Memory Triggers** with 85%+ accuracy, and **OAuth 2.1 team collaboration**. Works with **Claude Desktop, VS Code, Cursor, Continue, and 13+ AI applications**. <img width="240" alt="MCP Memory Service" src="https://github.com/user-attachments/assets/eab1f341-ca54-445c-905e-273cd9e89555" /> ## 🚀 Quick Start (2 minutes) ### 🆕 Latest Release: **v8.50.0** (Dec 09, 2025) **Fallback Quality Scoring - DeBERTa + MS-MARCO Hybrid System** - 🔄 **Multi-Model Fallback** - DeBERTa primary with MS-MARCO rescue for technical content (solves prose bias) - 📈 **Technical Content Rescue** - Expected 45-65% improvement for technical fragments (0.48 → 0.70-0.80) - ⚡ **Smart Performance** - Fast path (115ms, 40% of memories) for confident scores, full fallback (155ms, 60%) - 🎯 **Important Discovery** - MS-MARCO limitations documented (query-document relevance model, not quality classifier) - ✅ **Recommended Configuration** - Implicit signals only for technical corpora (DeBERTa prose bias confirmed) - 📊 **Complete Test Coverage** - 20/20 tests passing with comprehensive fallback logic validation **Previous Releases**: - **v8.49.0** - DeBERTa Quality Classifier (absolute quality assessment, eliminates self-matching bias) - **v8.48.4** - Cloudflare D1 Drift Detection Performance (10-100x faster queries, numeric comparison fix) - **v8.48.3** - Code Execution Hook Fix - 75% token reduction now working (fixed time_filter parameter, Python warnings, venv detection) - **v8.48.2** - HTTP Server Auto-Start & Time Parser Improvements (smart service management, "last N periods" support) - **v8.48.1** - Critical Hotfix - Startup Failure Fix (redundant calendar import removed, immediate upgrade required) - **v8.48.0** - CSV-Based Metadata Compression (78% size reduction, 100% sync success, metadata validation) - **v8.47.1** - ONNX Quality Evaluation Bug Fixes (self-match fix, association pollution, sync queue overflow, realistic distribution) - **v8.47.0** - Association-Based Quality Boost (connection-based enhancement, network effect leverage, metadata persistence) - **v8.46.3** - Quality Score Persistence Fix (ONNX scores in hybrid backend, metadata normalization) - **v8.46.2** - Session-Start Hook Crash Fix + Hook Installer Improvements (client-side tag filtering, isolated version metadata) - **v8.46.1** - Windows Hooks Installer Fix + Quality System Integration (UTF-8 console configuration, backend quality scoring) - **v8.45.3** - ONNX Ranker Model Export Fix (automatic model export, offline mode support, 7-16ms CPU performance) - **v8.45.2** - Dashboard Dark Mode Consistency Fixes (global CSS overrides,