Memento Protocol Enhanced
An enhanced wrapper around memento-mcp that adds sophisticated memory management capabilities inspired by the original ChatGPT memory design concepts.
š Features
š Protocol Memory System
Rule Enforcement Outside LLM: Protocols are enforced deterministically, not subject to model forgetfulness
YAML Configuration: Easy-to-edit protocol definitions
Auto Git Backup: Automatic version control before file modifications
Extensible Actions: Git operations, file system actions, API calls
šÆ Quality Management
Two-Stage Filtering: Heuristic + LLM validation for accuracy
Confidence Scoring: Tracks reliability of memories
Freshness Decay: Automatic aging and archival of old memories
Archival Tiers: Hot/Warm/Cold storage based on usage and age
š Enhanced Search (Hybrid Recall)
Multiple Strategies: Semantic vector, keyword matching, temporal relevance, confidence weighting
Hybrid Scoring: Combines multiple search approaches for better results
Quality Filtering: Filters results by confidence and relevance thresholds
Search Metadata: Detailed information about search process and results
š "Ask the Scribe" Synthesis Reports
Memory Synthesis: Combines related memories into coherent summaries
Insight Extraction: Identifies key patterns and connections
Confidence Tracking: Rates the reliability of synthesized information
Query-Focused: Tailored responses to specific questions
š§ Wrapper Architecture
Preserves Compatibility: Works as a drop-in replacement for memento-mcp
Upstream Safe: Doesn't fork memento-mcp, wraps it instead
Optional Features: All enhancements can be enabled/disabled
Graceful Fallbacks: Falls back to basic functionality if enhancements fail
š Quick Start
Installation
Basic Usage
Run Example
š Protocol System
Protocols are defined in YAML files and enforce rules automatically:
šÆ Quality Management
The quality system addresses six failure modes identified in basic memory systems:
Noise Accumulation: Filters low-quality information
Confidence Erosion: Tracks reliability over time
Retrieval Brittleness: Multiple search strategies for robustness
Temporal Confusion: Time-aware relevance scoring
Context Loss: Preserves rich metadata and relationships
Scale Degradation: Efficient archival and tier management
š Search Strategies
The hybrid search system combines multiple approaches:
Semantic Vector: Embedding-based similarity search
Keyword Matching: Exact term matching with scoring
Temporal Relevance: Recent memories weighted higher
Confidence Weighted: High-confidence memories prioritized
š Synthesis Reports
"Ask the Scribe" generates comprehensive reports by:
Multi-Strategy Search: Finds relevant memories using all search approaches
Quality Filtering: Removes low-confidence or irrelevant results
Insight Extraction: Identifies patterns and key information
Coherent Synthesis: Combines findings into readable summaries
Confidence Rating: Provides reliability assessment
šļø Architecture
The wrapper is built in distinct layers:
š§ Configuration
š MCP Server
The package includes a complete MCP server implementation:
š ļø Development
Requirements
Node.js 18+
Git (for protocol auto-backup)
Scripts
Adding Protocols
Create YAML file in
protocols/directoryDefine triggers, conditions, and actions
Protocol engine loads automatically
Extending Search
Add new search strategies in src/enhanced-search/index.js:
š¤ Integration
With HexTrackr
This wrapper was designed for integration with HexTrackr but works standalone:
With Other Projects
The wrapper preserves full memento-mcp compatibility:
šÆ Original Vision
This implementation realizes the original ChatGPT memory design vision:
"Some of our improvements with how we handle the searching and the semantics might actually be an improvement" - User feedback
The wrapper addresses fundamental limitations in basic memory systems while maintaining simplicity and compatibility.
Failure Modes Addressed
LLM compliance is unreliable ā Protocol enforcement outside LLM
Noisy memories from keyword scraping ā Two-stage filtering
Memory bloat & drift ā Freshness decay + archival tiers
Conflicting protocols ā Priority and scope management
Identity & grounding issues ā Stable IDs and linkage
Security & PII sprawl ā Secret scrubbing and access controls
Core Innovations
Hybrid Recall: Symbolic (SQL-like) + Vector (semantic) + Raw transcripts
Protocol Memory: Rules enforced deterministically, not stored as "memories"
Quality Pipeline: Heuristics ā LLM validation ā Confidence scoring
Archival Strategy: Hot/warm/cold tiers based on adjusted confidence
š License
MIT License - see LICENSE file for details.
š Links
memento-mcp - Core memory functionality
HexTrackr - Integration target project
Model Context Protocol - MCP specification
Enhanced memory management that learns, improves, and remembers.
This server cannot be installed
Related Resources
Related MCP Servers
- AsecurityAlicenseAqualityA custom Memory MCP Server that acts as a cache for Infrastructure-as-Code information, allowing users to store, summarize, and manage notes with a custom URI scheme and simple resource handling.Last updated -231MIT License
- AsecurityAlicenseAqualityA customized MCP memory server that enables creation and management of a knowledge graph with features like custom memory paths and timestamping for capturing interactions via language models.Last updated -104MIT License
- -securityFlicense-qualityAn MCP server that allows Claude and other LLMs to manage persistent memories across conversations through text file storage, enabling commands to add, search, delete and list memory entries.Last updated -517
- -securityAlicense-qualityEnhances the MCP memory server by implementing PouchDB for robust document storage and enabling the creation and management of a knowledge graph that captures interactions via language models.Last updated -4MIT License