Promptguard MCP Server: Install & Setup

Install Promptguard in Claude Code

Claude Code

claude mcp add --transport http promptguard https://mcp.glc-rag.hu/mcp

mcp.json (Claude Desktop, Cursor, VS Code, Windsurf)

{
  "mcpServers": {
    "promptguard": {
      "url": "https://mcp.glc-rag.hu/mcp"
    }
  }
}

Pass your API key, usually as an "Authorization: Bearer <token>" header (--header in Claude Code).

Layered prompt-injection checks for LLM hosts; 100 welcome credits on signup.

Probably not entirely. The listing mentions pay-per-call, credits or a subscription. Check the provider for current pricing.

Remote. Hosted by the provider. You connect with a URL, nothing to install.

Run: claude mcp add --transport http promptguard https://mcp.glc-rag.hu/mcp

Agent identity, scoped access, and proof across tools. Check authority before an agent acts.

Check text before your agent reads it: hidden instructions, invisible characters and links that leak data.

Notarize AI outputs, audits and agreements on the TON blockchain, with nothing stored server-side.

pollinations/model-context-protocol

Multimodal MCP server for generating images, audio, and text with no authentication required

I4cTime/quantum_ring

Quantum-inspired keyring for AI coding agents, securing secrets with superposition, entanglement, tunneling and teleportation.

Public phishing-domain feed: check a domain, search detections by brand, and pivot on campaigns and certs.