Install guide and config at curatedmcp.com

Tuteliq: Add Enterprise-Grade Content Moderation to Your AI Stack

When you expose Claude or Cursor to user-generated content, you inherit risk: grooming, fraud, deepfakes, self-harm signals, and synthetic CSAM. Tuteliq is a free MCP server that plugs real-time threat detection into your AI agent, catching 16+ categories of online harm across text, voice, image, video, and PDFs—all with structured, confidence-scored output you can act on programmatically.

What It Does

Tuteliq exposes 50 tools designed for child safety and trust & safety teams. It detects grooming, sextortion, romance scams, coercive control, radicalisation, gambling harm, and AI-generated synthetic content in a single request. Each detection returns:

• Per-message risk scores with confidence levels
• Age-calibrated thresholds (detection adapts to user age)
• Evidence-tagged categories (you see why it flagged something)
• Cross-endpoint amplification (correlate signals across multiple user messages)
• Country-aware crisis helplines (auto-route high-risk cases to local support)

Built-in GDPR compliance tools handle consent workflows, data export/deletion, audit logging, and breach reporting—critical if you're subject to KOSA, EU DSA, or local data-protection law. 27 languages, sub-second latency, and OAuth 2.1 auth mean you can ship this without custom token management.

How to Install

https://api.tuteliq.ai/mc

Add to your Claude Desktop config:

{
  "mcpServers": {
    "tuteliq": {
      "command": "https://api.tuteliq.ai/mc"
    }
  }
}

Authenticate via OAuth 2.1 PKCE on first use—no manual token handling required.

Real-World Use Cases

• Content moderation for user-facing chatbots: Detect grooming or social engineering in real-time before responses are sent, flag conversations for human review, and log evidence for compliance audits.
• Media safety in UGC platforms: Scan uploaded images and videos for deepfakes and synthetic CSAM; route high-confidence matches to NCMEC or equivalent authorities with structured evidence.
• Fraud prevention in financial AI agents: Identify romance scams, money-mule recruitment, and sextortion attempts in user messages; block transactions and escalate to fraud teams with per-message risk scores.

Full install guides for Claude Desktop, Cursor, Windsurf, and more at CuratedMCP.