Skip to content

Releases: gannonh/memento-mcp

v0.3.9

12 May 20:50
@gannonh gannonh
v0.3.9
be6dc36
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.
Compare
Choose a tag to compare
v0.3.9 Latest
Latest

[0.3.9] - 2025-05-08

Changed

  • Updated dependencies to latest versions:
    • @modelcontextprotocol/sdk from 1.8.0 to 1.11.0
    • axios from 1.8.4 to 1.9.0
    • dotenv from 16.4.7 to 16.5.0
    • eslint from 9.23.0 to 9.26.0
    • eslint-config-prettier from 10.1.1 to 10.1.3
    • glob from 11.0.1 to 11.0.2
    • openai from 4.91.1 to 4.97.0
    • tsx from 4.19.3 to 4.19.4
    • typescript from 5.8.2 to 5.8.3
    • vitest and @vitest/coverage-v8 from 3.1.1 to 3.1.3
    • zod from 3.24.2 to 3.24.4
    • @typescript-eslint/eslint-plugin and @typescript-eslint/parser from 8.29.0 to 8.32.0
Assets 2
Loading

v0.3.8 - Initial Release

03 Apr 14:00
@gannonh gannonh
v0.3.8
b9cb63a
Compare
Choose a tag to compare
v0.3.8 - Initial Release

Memento MCP v0.3.8: Initial Release

We're excited to announce the first public release of Memento MCP, a scalable, high-performance knowledge graph memory system for LLMs. This release provides persistent ontological memory for any LLM client that supports the Model Context Protocol (MCP).

Key Features

Core Functionality

  • Knowledge Graph Memory: Store and manage entities and relations with full version history
  • Semantic Search: Find semantically related entities based on meaning rather than just keywords
  • Temporal Awareness: Track complete history with point-in-time graph retrieval
  • Confidence Decay: Relations automatically decay in confidence over time

Data Architecture

  • Neo4j Backend: Unified storage for both graph data and vector embeddings
  • Vector Search: Built-in semantic search using vector embeddings
  • Rich Metadata: Support for custom metadata on both entities and relations
  • Enhanced Relations: Relations include strength, confidence, and custom metadata

Integration

  • Model Context Protocol: Full integration with MCP-compatible LLM clients
  • Claude Desktop Support: Ready-to-use configuration for Claude Desktop
  • Cursor Integration: Support for the Cursor IDE
  • GitHub Copilot: Compatible with GitHub Copilot

Developer Experience

  • Docker Support: Easy setup with Docker Compose
  • CLI Utilities: Tools for database management and diagnostics
  • Comprehensive Documentation: Detailed setup and usage instructions
  • Extensive Testing: High test coverage and reliability

Requirements

  • Node.js 20.0.0 or higher
  • Neo4j 5.13+ (for vector search capabilities)
  • OpenAI API key (for semantic embeddings)

Documentation

License

MIT

Loading