# Hypermnesia MCP

> Hypermnesia MCP is an open-source MCP server for knowledge and memory. Persistent, local-first memory for Claude Code with hybrid retrieval and cited sources. It runs locally on your machine and is started by your AI client.

- **Kind:** MCP server
- **Pricing:** Open source
- **Platforms:** CLI, self-hosted
- **Website:** https://ai-architect.tools
- **Repository:** https://github.com/cdeust/cortex
- **MCP Registry name:** io.github.cdeust/hypermnesia-mcp
- **Categories:** knowledge and memory, developer tools
- **Listing updated:** 2026-11-14

## About

Hypermnesia is the MCP server form of the Cortex memory project. The README counts 36 implemented neuroscience-inspired mechanisms, 57 memory tools, 11 lifecycle hooks and a self-curating per-project wiki, all local and MIT licensed. Mechanisms named on the site include spreading activation, consolidation cycles and microglial pruning.

Retrieval fuses five signals, and the site reports 97.8 percent Recall@10 on LongMemEval-S for Cortex v4.20.0. A stated principle is that every claim ties back to a memory, file, commit or citation, and verification is deterministic, with no model judging another model.

It installs from PyPI over stdio. The maker is a one-person consultancy that also sells code migrations at fixed prices after a paid pilot; the memory server itself is one of several open-source components alongside a read-only visualiser and a session optimiser.

## Features

- **Neuroscience-based memory:** Implements 36 documented mechanisms such as consolidation and pruning.
- **Hybrid retrieval:** Fuses five retrieval signals per the maker.
- **Lifecycle hooks:** 11 hooks tie memory into agent sessions.
- **Per-project wiki:** Curates a wiki for each project automatically.
- **Cited facts:** Claims link back to a memory, file, commit or citation.
- **Local and open source:** MIT licensed and runs on your machine.

## Good for

- Let an agent remember decisions across Claude Code sessions
- Keep per-project knowledge without a hosted service
- Trace an agent's statement back to its source memory

## Install

### Claude Code (Terminal)

```sh
claude mcp add hypermnesia-mcp -- uvx hypermnesia-mcp
```

### Claude Desktop (claude_desktop_config.json)

```json
{
  "mcpServers": {
    "hypermnesia-mcp": {
      "command": "uvx",
      "args": [
        "hypermnesia-mcp"
      ]
    }
  }
}
```

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

```json
{
  "mcpServers": {
    "hypermnesia-mcp": {
      "command": "uvx",
      "args": [
        "hypermnesia-mcp"
      ]
    }
  }
}
```

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

```json
{
  "servers": {
    "hypermnesia-mcp": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "hypermnesia-mcp"
      ]
    }
  }
}
```

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

```json
{
  "mcpServers": {
    "hypermnesia-mcp": {
      "command": "uvx",
      "args": [
        "hypermnesia-mcp"
      ]
    }
  }
}
```

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

```json
{
  "mcpServers": {
    "hypermnesia-mcp": {
      "command": "uvx",
      "args": [
        "hypermnesia-mcp"
      ]
    }
  }
}
```

## FAQ

### Does it run locally?

Yes. The README describes it as local-first and open source under MIT.

### How is it installed?

From PyPI as hypermnesia-mcp, over stdio.

### Is the benchmark figure independent?

The 97.8 percent Recall@10 on LongMemEval-S is the maker's own reported result.

### Is there a visualiser?

A separate read-only tool, hypermnesia-mcp-viz, is listed.

### Is Hypermnesia MCP free?

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

### How do I add Hypermnesia MCP to Claude Code?

Run this command in your terminal: claude mcp add hypermnesia-mcp -- uvx hypermnesia-mcp

### What are the best alternatives to Hypermnesia MCP?

The closest alternatives to Hypermnesia MCP are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

## Alternatives

- [REA](https://ai-powered.com/mcp/rea): REA is an open-source MCP server for developer tools.
- [Codebase Memory](https://ai-powered.com/mcp/codebase-memory): Codebase Memory is an open-source MCP server for knowledge and memory.
- [GitHub](https://ai-powered.com/mcp/github-2): GitHub is an open-source MCP server for developer tools.
- [Serena MCP: the IDE for your agent](https://ai-powered.com/mcp/serena-mcp-the-ide-for-your-agent): Serena MCP: the IDE for your agent is an open-source MCP server for developer tools.
- [XERJ](https://ai-powered.com/mcp/xerj): XERJ is an open-source MCP server for knowledge and memory.
- [Ouroboros](https://ai-powered.com/mcp/ouroboros): Ouroboros is an open-source MCP server for developer tools.
- [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.

---
Source: https://ai-powered.com/mcp/hypermnesia-mcp
