MCP Server

io.github.ipezygj/agent-guard MCP Server

Name: io.github.ipezygj/agent-guard MCP Server
Author: io.github.ipezygj

io.github.ipezygj/agent-guard

What is the io.github.ipezygj/agent-guard MCP server?

Safety checks an agent runs before it acts: package malware, destructive command, secret leak.

Copy-paste configuration for popular MCP clients.

transport: stdio

Config generated by PluginBench — verify against the source before use.

Claude Desktop

~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "agent-guard": {
      "command": "uvx",
      "args": [
        "agent-tripwire"
      ]
    }
  }
}

Cursor

~/.cursor/mcp.json

{
  "mcpServers": {
    "agent-guard": {
      "command": "uvx",
      "args": [
        "agent-tripwire"
      ]
    }
  }
}

Windsurf

~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "agent-guard": {
      "command": "uvx",
      "args": [
        "agent-tripwire"
      ]
    }
  }
}

VS Code

.vscode/mcp.json

{
  "servers": {
    "agent-guard": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "agent-tripwire"
      ]
    }
  }
}

Claude Code

claude mcp add agent-guard -- uvx agent-tripwire

Active

Verify a number before an agent asserts it — evals, backtests (Deflated Sharpe), signed receipts.

Python

MIT

Statistical checks an agent runs before trusting an AI eval number (is #1 real, judge bias, more).