MCP.so
登录

Memory Quality Mcp

@ladyiceberg

关于 Memory Quality Mcp

A Claude Code MCP plugin that audits and cleans up your AI memory.

基本信息

分类

记忆与知识

传输方式

stdio

发布者

ladyiceberg

提交者

ladyiceberg z

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "memory-quality": {
      "command": "uvx",
      "args": [
        "memory-quality-mcp"
      ],
      "env": {
        "OPENAI_API_KEY": "your-key-here"
      }
    }
  }
}

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is Memory Quality Mcp?

Memory Quality Mcp audits and cleans up Claude Code’s auto-saved memories. It provides a health check with four tools that scan, score, clean, and visualize your memory store. It supports multiple LLM providers and auto-detects language from system locale.

How to use Memory Quality Mcp?

Invoke the provided tools (memory_audit(), memory_report(), memory_cleanup(), memory_dashboard()) after installation. A demo mode is available that requires no real memory files.

Key features of Memory Quality Mcp

  • Instant memory scan with no LLM cost (memory_audit())
  • Four-dimension scoring: Importance, Recency, Credibility, Accuracy
  • Conflict detection and safe cleanup with dry-run preview
  • Auto-backup of cleaned memories to .trash/ folder
  • Visual health dashboard with scores and recommendations
  • Supports OpenAI, Anthropic, Kimi, and MiniMax providers

Use cases of Memory Quality Mcp

  • Audit a bloated memory store for stale or junk entries
  • Identify conflicting preferences saved over many sessions
  • Clean up one-off remarks that were over-interpreted as facts
  • Generate a health report to understand memory quality at a glance

FAQ from Memory Quality Mcp

What does Memory Quality Mcp do?

It scans and cleans up the automatic memories that Claude Code saves from conversations, helping you remove stale, conflicting, or junk entries.

Do I need real memory files to try it?

No. You can use demo mode first—no real memory files are required.

What tools are available?

Four tools: memory_audit() (instant scan), memory_report() (scoring + conflict detection), memory_cleanup() (safe cleanup with backup), and memory_dashboard() (visual health dashboard).

Which LLM providers are supported?

OpenAI, Anthropic, Kimi, and MiniMax are explicitly mentioned.

Is the cleanup operation safe?

Yes. It offers a dry-run preview and automatically backs up memories to a .trash/ folder before performing any cleanup.

评论

记忆与知识 分类下的更多 MCP 服务器