@safenest/mcp

SafeNest MCP Server

MCP server for SafeNest - AI-powered child safety tools for Claude

API Docs • Dashboard • Discord

What is this?

SafeNest MCP Server brings AI-powered child safety tools directly into Claude, Cursor, and other MCP-compatible AI assistants. Ask Claude to check messages for bullying, detect grooming patterns, or generate safety action plans.

Available Tools

Tool	Description
detect_bullying	Analyze text for bullying, harassment, or harmful language
detect_grooming	Detect grooming patterns and predatory behavior in conversations
detect_unsafe	Identify unsafe content (self-harm, violence, explicit material)
analyze	Quick comprehensive safety check (bullying + unsafe)
analyze_emotions	Analyze emotional content and mental state indicators
get_action_plan	Generate age-appropriate guidance for safety situations
generate_report	Create incident reports from conversations

Installation

Claude Desktop

Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json on Mac):

{
  "mcpServers": {
    "safenest": {
      "command": "npx",
      "args": ["-y", "@safenest/mcp"],
      "env": {
        "SAFENEST_API_KEY": "your-api-key"
      }
    }
  }
}

Cursor

Add to your Cursor MCP settings:

{
  "mcpServers": {
    "safenest": {
      "command": "npx",
      "args": ["-y", "@safenest/mcp"],
      "env": {
        "SAFENEST_API_KEY": "your-api-key"
      }
    }
  }
}

Global Install

npm install -g @safenest/mcp

Then run:

SAFENEST_API_KEY=your-api-key safenest-mcp

Usage Examples

Once configured, you can ask Claude:

Bullying Detection

"Check if this message is bullying: 'Nobody likes you, just go away'"

Response:

## ⚠️ Bullying Detected

**Severity:** 🟠 Medium
**Confidence:** 92%
**Risk Score:** 75%

**Types:** exclusion, verbal_abuse

### Rationale
The message contains direct exclusionary language...

### Recommended Action
`flag_for_moderator`

Grooming Detection

"Analyze this conversation for grooming patterns..."

Quick Safety Check

"Is this message safe? 'I don't want to be here anymore'"

Emotion Analysis

"Analyze the emotions in: 'I'm so stressed about school and nobody understands'"

Action Plan

"Give me an action plan for a 12-year-old being cyberbullied"

Incident Report

"Generate an incident report from these messages..."

Get an API Key

• Go to safenest.app
• Create an account
• Generate an API key
• Add it to your MCP config

Requirements

• Node.js 18+
• SafeNest API key

Best Practices

Message Batching

The bullying and unsafe content tools analyze a single text field per request. If you're analyzing a conversation, concatenate a sliding window of recent messages into one string rather than sending each message individually. Single words or short fragments lack context for accurate detection and can be exploited to bypass safety filters.

The grooming tool already accepts a messages[] array and analyzes the full conversation in context.

PII Redaction

Enable PII_REDACTION_ENABLED=true on your SafeNest API to automatically strip emails, phone numbers, URLs, social handles, IPs, and other PII from detection summaries and webhook payloads. The original text is still analyzed in full — only stored outputs are scrubbed.

Support

• API Docs: api.safenest.dev/docs
• Discord: discord.gg/7kbTeRYRXD
• Email: [e-mail]

License

MIT License - see LICENSE for details.

_{Built with care for child safety by the SafeNest team}