Agent Output Guard

Overview

Agent Output Guard is a Model Context Protocol server that validates and verifies data from other agents before acting on it, with zero LLM costs. By implementing MCP, it lets AI assistants call validation tools directly rather than requiring manual data transfer between applications.

The server runs locally via the npm package @agenson-horrowitz/agent-output-guard-mcp, communicating over stdio without requiring any credentials. Version v1.0.8 is published under the MIT license, with source code available at agenson-tools/agent-output-guard-mcp on GitHub.

Agent Output Guard is listed in the AI & Agents category on mcp.site and works with MCP clients including Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, and Zed. If you maintain this project, claim this listing to verify ownership and keep information current.

Installation

// claude_desktop_config.json
{
  "mcpServers": {
    "agent_output_guard": {
      "command": "npx",
      "args": [
        "-y",
        "@agenson-horrowitz/agent-output-guard-mcp"
      ]
    }
  }
}