# emem

emem is an MCP server that gives AI agents shared signed memory. It lets them exchange verifiable observations and spatial data without referential drift.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · 1175 installs
- Author: vortxai
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://smithery.ai/server/vortxai/emem
- Updated: 2026-05-01
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/emem

## What it does

The `emem` server provides a shared memory layer for autonomous agents interacting with physical environments. Instead of exchanging compressed summaries, models read byte-identical observations. Each record receives a single content address. Every fact includes a cryptographic signature proving how it was generated. Read requests require no account or API keys.

The server exposes sixteen tools. Calling `emem_locate` turns a geographic location into a vendor-neutral `cell64` address. Tool `emem_recall` retrieves verified attestations from storage. When separate sources submit conflicting measurements for the same coordinate, `emem_memory_contradictions` flags the dispute with a severity score from 0 to 1. The `emem_guard_verdict` handler checks generated drafts against citations before transmission.

Similarity search runs through `emem_find_similar` using exact cosine math or hamming popcounts. For receipt checks, `emem_verify_receipt` reconstructs preimages according to version v1 or v2 rules. Verifying signatures server-side prevents unvetted claims from polluting the shared agent graph.

## Requirements
- Remote server: `https://emem--vortxai.run.tools`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http emem https://emem--vortxai.run.tools
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add emem --url https://emem--vortxai.run.tools
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "emem": {
            "url": "https://emem--vortxai.run.tools"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "emem": {
            "type": "http",
            "url": "https://emem--vortxai.run.tools"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http emem https://emem--vortxai.run.tools
```

## When to use it
- Ground fuzzy location descriptions into canonical `cell64` addresses
- Verify raw observation records using cryptographic receipt envelopes
- Detect conflicting evidence from separate sources with severity scoring
- Validate outgoing citations with `emem_guard_verdict` before sending

## Example prompts
- Locate this ground position with `emem_locate` and retrieve stored facts.
- Inspect conflicting reports for this place using `emem_memory_contradictions`.
- Run `emem_guard_verdict` on the drafted text to catch citation drift.

## Similar
- [memoweft](https://affpapa.org/en/nevibecode/mcp/memoweft.md) — MCP server · ★ 6 repo stars
- [V3 Memory Unification](https://affpapa.org/en/nevibecode/skills/v3-memory-unification.md) — skill · 1100 installs
- [VelesDB Memory](https://affpapa.org/en/nevibecode/mcp/velesdb-memory.md) — MCP server · ★ 105 repo stars
- [SAME - Stateless Agent Memory Engine](https://affpapa.org/en/nevibecode/mcp/same-stateless-agent-memory-engine.md) — MCP server · ★ 24 repo stars
- [Holographic Memory](https://affpapa.org/en/nevibecode/mcp/holographic-memory.md) — MCP server · ★ 23 repo stars
- [mnemo](https://affpapa.org/en/nevibecode/mcp/mnemo.md) — MCP server · ★ 13 repo stars

---
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
