広告掲載
MCP.so
ログイン

Optimized Memory MCP Server v2

@AgentWong

リポジトリ

基本情報

カテゴリ

メモリとナレッジ

トランスポート

stdio

作者

AgentWong

Optimized Memory MCP Server v2 について

This is a personal project to test Claude AI's ability to self-write an MCP Server code for its own use.

#mcp#server#claude-ai

Optimized Memory MCP Server v2 の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "optimized-memory-mcp-serverv2": {
      "command": "uvx",
      "args": [
        "run",
        "python",
        "-m",
        "src.main"
      ]
    }
  }
}

Optimized Memory MCP Server v2 のツール

ツール一覧はまだありません

この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。

上のボタンで一時的なサンドボックスでこのサーバーを実行し、ツール一覧を取得できます。

概要

What is Optimized Memory MCP Server v2?

Optimized Memory MCP Server v2 is a high-performance Python-based Model Context Protocol (MCP) server implementation optimized for Claude Desktop integration. It provides efficient memory management and robust infrastructure component tracking capabilities, using SQLite for persistent storage with connection pooling.

How to use Optimized Memory MCP Server v2?

Clone the repository, set up a Python environment (Python 3.13.1+), install dependencies with pip install -r requirements.txt, configure the database URL via the DATABASE_URL environment variable, run database migrations with alembic upgrade head, then start the server with uvx run python -m src.main. In Claude Desktop, set the MCP server URL to http://localhost:8000 and enable the MCP protocol in settings.

Key features of Optimized Memory MCP Server v2

  • MCP resources for entity, provider, and Ansible collection management
  • MCP tools for entity creation, observation, and provider registration
  • Full-text search across tracked infrastructure components
  • SQLite database with connection pooling and automatic cleanup
  • Comprehensive error handling and extensive logging
  • Full Claude Desktop compatibility via FastMCP

Use cases of Optimized Memory MCP Server v2

  • Tracking infrastructure components and their relationships in a centralized store
  • Managing Ansible collections and provider resources through MCP tools
  • Performing full-text search on entities, observations, and metadata
  • Integrating with Claude Desktop for AI-assisted infrastructure analysis

よくある質問

1

What are the runtime requirements?

Python 3.13.1 or higher, SQLite 3.x, and the uvx server. Dependencies are listed in `requirements.txt`.

2

How do I start the server?

Run `uvx run python -m src.main` after installing dependencies, configuring the database URL, and applying migrations with `alembic upgrade head`.

3

How do I connect Claude Desktop to this server?

Set the MCP server URL in Claude Desktop to `http://localhost:8000` and enable the MCP protocol in Claude settings. Verify the connection with `curl http://localhost:8000/health`.

4

How is data stored?

Data is stored persistently in a SQLite database. The database URL is configured via the `DATABASE_URL` environment variable (e.g., `sqlite:///path/to/db.db`). Migrations are managed with Alembic.

5

Is this project still maintained?

No. The README indicates the project has been archived due to faulty project specifications and AI direction that led to endless looping behavior.

コメント

「メモリとナレッジ」の他のコンテンツ

PLUR

plur-ai

AI agents start every session with amnesia — you re-explain the project, repeat your preferences, and correct the same mistakes over and over. PLUR gives them a memory that persists. Your agent's corrections, preference

226

agentleFS

agentleFS

One permissioned folder your team and your AI agents both log into.

ctxfile

ctxfile

Local-first MCP server that snapshots your project's working state into one context object. Open-core, privacy-first.

2

Takibi Base

Deian Isac

Ask your Takibi Base collections anything; get answers with cited passages and a support score.

LoreKeeper RPG (lorekeeper.com)

LoreKeeper Ltd

Mess in, magic out: pour in your D&D and TTRPG notes and LoreKeeper keeps them as a private store for your world. Ask Claude or ChatGPT about it, or build a sourcebook.

広告掲載