# Memwyre

External shared memory layer for AI agents. Retains knowledge and facts across multiple coding sessions via MCP.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 27
- Author: memwyreai
- License: Apache-2.0 · policy=link-only · description only; the file stays with the author
- Source: https://github.com/memwyreai/Memwyre
- Updated: 2026-09-16
- HTML: https://affpapa.org/en/nevibecode/mcp/memwyre

## What it does

The server provides a shared long-term memory store for developer assistants. It groups user notes by workspace and lowers the priority of outdated details using an Ebbinghaus logarithmic curve. This avoids token waste in long prompts.

Under the hood, the MCP layer surfaces three core tools: `save_memory`, `search_memory`, and `get_document`. The architecture couples a FastAPI application with Celery task queues, SQLite or PostgreSQL for records, and vector backends like Pinecone or ChromaDB.

Connecting through MCP requires network access to `https://server.memwyre.tech/mcp/`. Auto-captured entries route to an inbox first, so you can discard noisy facts before vector indexing commits them.

## Requirements
- Remote server: `https://server.memwyre.tech/mcp/`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http memwyre https://server.memwyre.tech/mcp/
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add memwyre --url https://server.memwyre.tech/mcp/
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "memwyre": {
            "url": "https://server.memwyre.tech/mcp/"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "memwyre": {
            "type": "http",
            "url": "https://server.memwyre.tech/mcp/"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http memwyre https://server.memwyre.tech/mcp/
```

## When to use it
- persist developer context across distinct chat sessions
- locate recorded technical decisions using `search_memory`
- isolate facts into project-specific containers
- discard obsolete preferences to prevent token bloat

## Example prompts
- Store this database migration plan in Memwyre using `save_memory`
- Find our earlier API design notes with `search_memory`
- Fetch the linked specification text via `get_document`

## 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

---
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