# neural-memory

This MCP server keeps agent memories between sessions and retrieves related facts through a graph. Use it when project context must carry across runs.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 242
- Author: nhadaututtheky
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/nhadaututtheky/neural-memory
- Updated: 2026-10-23
- HTML: https://affpapa.org/en/nevibecode/mcp/neural-memory

## What it does

neural-memory stores facts, decisions, errors, preferences, and other records as connected memory items. Its recall process follows associations between them instead of relying only on matching words. The MCP tools cover saving memories, recalling context, and checking brain health, while session and maintenance work runs in the background.

The server runs locally without a vector database or API calls. It initializes itself on first use, so there is no separate `init` step. Documents in PDF, DOCX, PPTX, HTML, JSON, XLSX, and CSV can be ingested as lasting knowledge. Device-to-device sync uses a Cloudflare Worker deployed in your own account.

The project lists Python 3.11+, so that version should be checked before setup. Cloud sync also needs your own Cloudflare account and D1 database. Local recall does not depend on either service.

## Requirements
- Needs: Node.js (commands run via npx)

## Install

### Claude Code (native platform)

**Platform CLI**
```bash
claude mcp add neural-memory -- npx -y neural-memory-mcp
```

### Codex

**Platform CLI**
```bash
codex mcp add neural-memory -- npx -y neural-memory-mcp
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "neural-memory": {
            "command": "npx",
            "args": [
                "-y",
                "neural-memory-mcp"
            ]
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "neural-memory": {
            "type": "stdio",
            "command": "npx",
            "args": [
                "-y",
                "neural-memory-mcp"
            ]
        }
    }
}
```

### Gemini CLI

**Platform CLI**
```bash
gemini mcp add neural-memory npx -y neural-memory-mcp
```

## When to use it
- keep project decisions across agent sessions
- trace related memories through cause-and-effect links
- load knowledge from PDF and office documents
- inspect the health of stored memory
- sync one memory store across devices

## Example prompts
- Remember that this project uses PostgreSQL for its database decision.
- Recall the related context behind the authentication changes.
- Check the memory health and suggest fixes for the issues you find.

## Similar
- [REA](https://affpapa.org/en/nevibecode/mcp/rea.md) — MCP server · ★ 386989
- [GitHub](https://affpapa.org/en/nevibecode/mcp/github.md) — MCP server · ★ 33918
- [Netdata](https://affpapa.org/en/nevibecode/mcp/netdata.md) — MCP server · ★ 81115
- [browser-use](https://affpapa.org/en/nevibecode/mcp/browser-use.md) — MCP server · ★ 119160
- [Codebase Memory](https://affpapa.org/en/nevibecode/mcp/codebase-memory.md) — MCP server · ★ 47392
- [mcp-server](https://affpapa.org/en/nevibecode/mcp/mcp-server.md) — MCP server · ★ 43872

From the same set: [nhadaututtheky/neural-memory](https://affpapa.org/en/nevibecode/by/nhadaututtheky/neural-memory.md) — 4 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md