I

Injection Guard

Verified Integration
Author: @Hosein-AbdollahiCategory: ServerApplication
JSON-RPC 2.0
Protocol Standard
Sub-second
Execution Latency
Active
Operational Status

Client Configuration

— Connect Injection Guard to Claude Desktop or Cursor in seconds
{
  "mcpServers": {
    "injection-guard": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-injection-guard"
      ],
      "env": {}
    }
  }
}
Paste into ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows).
Architecture & Capabilities

System Overview

Blocks indirect prompt injection in AI agents by tracking data provenance rather than text patterns.

Indexed Date

7/23/2026

License

Open Source

Protocol Layer

stdio / SSE RPC

Frequently Asked Questions

Architecture and operational details for Injection Guard

Injection Guard functions as a standard stdio MCP server. It can be easily integrated with any MCP-compliant client, including Claude Desktop, Cursor, or Continue. You simply configure your client to run the `server.py` script, and it will begin monitoring and protecting your agent's interactions.

Related MCP Servers

Browse all servers →
C
@carboneio
Carbone

Empower AI assistants to generate and convert documents, manage templates, and automate document workflows efficiently.

ClaudeSearchDatabase+65 FAQs
Learn more
C
@sebastienrousseau
CloudCDN

Provides a multi-tenant, AI-native Content Delivery Network deployable on Cloudflare, featuring sub-100ms TTFB, AI agent controllability, and comprehensive accessibility.

ClaudeSearchDatabase+65 FAQs
Learn more

Deploy a Model Context Protocol server on Cloudflare Workers without requiring authentication.

ClaudeSearchDatabase+55 FAQs
Learn more