# Clotr Guard

Clotr Guard is an MCP server for Claude Code that scans text for credentials and personal details before it is sent. Hooks can check prompts and file reads.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 9 repo stars
- Author: clotr-app
- License: AGPL-3.0 · policy=link-only · description only; the file stays with the author
- Source: https://github.com/clotr-app/clotr/tree/HEAD/apps/guard
- Updated: 2026-10-29
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/clotr-guard

## What it does

Clotr Guard scans text on the computer running the AI client. Hooks can pause prompts and file reads, while the MCP tools run checks when an assistant calls them. Medium- and high-severity results pause a prompt and show the finding category with a masked sample. Low-severity results pass with a brief message on stderr.

The server communicates through standard input and output, and does not send scanned text to Clotr. Node.js 20 or newer is required. A paused prompt can be sent again within two minutes; a fresh finding pauses it again. Shell-command scanning is a preview feature, and an unrecognised command can still print text.

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

## Install

### Claude Code (native platform)

**Platform CLI**
```bash
claude mcp add clotr-guard -- npx -y clotr-guard
```

### Codex

**Platform CLI**
```bash
codex mcp add guard -- npx -y clotr-guard
```

### Cursor

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

### GitHub Copilot

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

### Gemini CLI

**Platform CLI**
```bash
gemini mcp add guard npx -y clotr-guard
```

## When to use it
- scan a prompt for credentials before sending it
- spot personal details in file reads
- pause prompts that contain sensitive findings
- run a text check through an assistant-called MCP tool

## Example prompts
- Scan this prompt for credentials and personal details.
- Check this file's contents before using them.
- Run Clotr Guard on this text and show masked findings.

## Similar
- [n8n-credentials-and-security-official](https://affpapa.org/en/nevibecode/skills/n8n-credentials-and-security-official.md) — skill · 2500 installs
- [secret-hygiene](https://affpapa.org/en/nevibecode/skills/secret-hygiene.md) — skill · 2200 installs
- [credentials](https://affpapa.org/en/nevibecode/skills/credentials.md) — skill · 1600 installs
- [secrets](https://affpapa.org/en/nevibecode/skills/ulpi-io-secrets.md) — skill · 148 installs
- [safety-guard](https://affpapa.org/en/nevibecode/skills/safety-guard.md) — skill · 6200 installs
- [1password](https://affpapa.org/en/nevibecode/mcp/1password.md) — MCP server · ★ 22 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
