# Memento

> Memento is an open-source MCP server for knowledge and memory. Give agents a local memory with four memory types, hybrid search and TTL-based forgetting. It runs locally on your machine and is started by your AI client.

- **Kind:** MCP server
- **Pricing:** Open source
- **Platforms:** self-hosted, CLI, Linux, macOS, Windows
- **Repository:** https://github.com/jee1/memento
- **MCP Registry name:** io.github.jee1/memento-mcp-server
- **Categories:** [knowledge and memory](https://ai-powered.com/categories/knowledge-and-memory)
- **Listing updated:** 2026-11-18

## About

Memento stores what an agent learns between sessions in a local SQLite database. Memories are typed as working, episodic, semantic or procedural through the type argument of remember, and recalled with a hybrid of full-text and vector search plus tag filters. Memories that get used are reinforced, stale ones are cleaned up by a forgetting policy, and similar memories are linked so key context can be restored as an anchor in the next conversation.

The server registers 22 tools, but tools/list shows only four by default: recall, remember, memory_injection and feedback. The rest still work when called, and setting MEMENTO_TOOLSET=full lists them all. Procedural memories have diff and rollback, and a case directive feature injects only the relevant instructions into Claude Code, Codex or Cursor prompts within a fixed budget.

Install as a Claude Code plugin or run npx memento-mcp-server@latest with a DB_PATH. An HTTP server adds a browser dashboard and memory graph, protected by session cookies and bearer or API-key headers. The README is mainly in Korean with an English edition, says Memento sends no usage data out, and is licensed MIT.

## Features

- **Four memory types:** Working, episodic, semantic and procedural memories are chosen when saving.
- **Hybrid recall:** Combines FTS5 text search, vector similarity and tag filters.
- **Reinforce and forget:** Used memories are strengthened and unused ones are removed by a TTL policy.
- **Procedural versioning:** Procedural memories support diff and rollback.
- **Directive injection:** Adds only relevant case instructions to each prompt within a fixed budget.
- **Dashboard and graph:** An HTTP server serves an admin dashboard and memory graph.

## Good for

- Keep project decisions available across separate chat sessions
- Store a learned procedure and roll it back if it goes wrong
- Restore key context at the start of a new conversation
- Share memory between several agents through the HTTP server
- Inject project-specific rules into coding assistant prompts

## Install

### Claude Code (Terminal)

```sh
claude mcp add memento-mcp-server -- npx -y memento-mcp-server
```

### Claude Desktop (claude_desktop_config.json)

```json
{
  "mcpServers": {
    "memento-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "memento-mcp-server"
      ]
    }
  }
}
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "memento-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "memento-mcp-server"
      ]
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "memento-mcp-server": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "memento-mcp-server"
      ]
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "memento-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "memento-mcp-server"
      ]
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "memento-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "memento-mcp-server"
      ]
    }
  }
}
```

## Tools

- `recall`: Search stored memories.
- `remember`: Save a memory with a type, tags and importance.
- `memory_injection`: Inject relevant stored context into a prompt.
- `feedback`: Give feedback on recalled memories.

## FAQ

### How do I install it?

In Claude Code use the plugin marketplace commands, or run npx memento-mcp-server@latest with a DB_PATH in your MCP config.

### Why do I see only four tools?

tools/list exposes recall, remember, memory_injection and feedback by default; set MEMENTO_TOOLSET=full to list all 22.

### Where is data stored?

In a local SQLite file at the path you set; the Claude Code plugin keeps it in its plugin data folder.

### Does it send usage data?

The README says it does not send usage information out.

### What is the licence?

MIT.

### Is Memento free?

Yes. Memento is open source, so you can use and self-host it for free.

### How do I add Memento to Claude Code?

Run this command in your terminal: claude mcp add memento-mcp-server -- npx -y memento-mcp-server

### What are the best alternatives to Memento?

The closest alternatives to Memento are Codebase Memory, XERJ, Claude Code Ultimate Guide, Nautilus Compass Memory and NeuroStack.

## Where this comes from

- Description: written from the project's README.
- Install steps: from the official MCP Registry entry.

## Alternatives

- [Codebase Memory](https://ai-powered.com/mcp/codebase-memory): Codebase Memory is an open-source MCP server for knowledge and memory.
- [XERJ](https://ai-powered.com/mcp/xerj): XERJ is an open-source MCP server for knowledge and memory.
- [Claude Code Ultimate Guide](https://ai-powered.com/mcp/claude-code-ultimate-guide): Claude Code Ultimate Guide is an open-source MCP server for ai models and llm tooling.
- [Nautilus Compass Memory](https://ai-powered.com/mcp/nautilus-compass-memory): Nautilus Compass Memory is an open-source MCP server for knowledge and memory.
- [NeuroStack](https://ai-powered.com/mcp/neurostack): NeuroStack is a freemium MCP server for knowledge and memory.
- [note1](https://ai-powered.com/mcp/note1): note1 is an MCP server for productivity.
- [Digitalnz](https://ai-powered.com/mcp/digitalnz): Digitalnz is a freemium MCP server for knowledge and memory.
- [Notion_connect](https://ai-powered.com/mcp/notion-connect): Notion_connect is a freemium MCP server for knowledge and memory.

---
Source: https://ai-powered.com/mcp/memento-2
