MCP Catalogs
Homenode9-proxy screenshot

node9-proxy

by node9-ai·180·Score 48

Node9 is an execution security layer for AI agents that provides governance, monitoring, and audit logging through MCP integration.

securitydeveloper-toolsai-llm
18
Forks
0
Open issues
this month
Last commit
2d ago
Indexed

Overview

Node9 serves as a security gateway between AI agents and their tools, offering discovery of past agent activities, real-time protection against risky commands, and comprehensive review capabilities. It works with various AI environments including Claude Code, Gemini CLI, Cursor, and any MCP server. The tool provides deterministic 'Sudo' governance with customizable shields for different services like databases, cloud platforms, and filesystem operations. Its MCP gateway feature allows transparent wrapping of existing MCP servers while intercepting and monitoring all tool calls.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

you:Monitor and govern AI agent behavior to prevent unauthorized access to credentials or sensitive data
you:Provide real-time blocking of dangerous commands like 'rm -rf', 'DROP TABLE', or force git pushes
you:Audit and review AI agent activities over time windows to analyze patterns, costs, and potential security issues
you:What is Node9?
you:How does Node9 work with MCP servers?

When to choose this

Choose Node9 when you need enterprise-grade security governance for AI agents, especially if you work with sensitive systems and require audit trails of all agent actions.

When NOT to choose this

Don't choose Node9 if you need a lightweight solution with minimal overhead, or if you're only working with non-critical systems where comprehensive governance isn't required.

Tools this server exposes

12 tools extracted from the README
  • scan

    Scan past AI sessions for credential leaks, agent loops, and secrets on disk

  • monitor

    Open an interactive dashboard for real-time activity and security alerts

  • report

    Generate period-windowed reports of agent activity, costs, and security incidents

  • blast

    Show what an AI agent can reach on the system - files, credentials, environment

  • tail

    Live stream of every tool call in text format

  • sessions

    Display session history with prompts, tool traces, costs, and snapshots

  • dlp

    Find credential leaks in Claude's response text

  • mask

    Redact plaintext secrets from local session history files

  • shield

    Manage security shields - rule packs for different services

  • mcp

    Manage MCP server configurations and tool pinning

  • undo

    Revert AI file edits using git snapshots

  • init

    Auto-wire Node9 with Claude Code, Gemini CLI, Cursor, Codex, and MCP servers

Comparable tools

deep-and-darkagent-monitorllm-guard

Installation

Install via npm or Homebrew:

# npm (any platform)
npm install -g node9-ai

# macOS / Linux
brew tap node9-ai/node9 && brew install node9

Initialize with:

node9 init     # auto-wires Claude Code, Gemini CLI, Cursor, Codex, MCP servers
node9 doctor   # verify everything is wired correctly

For Claude Desktop MCP integration:

{
  "mcpServers": {
    "node9": {
      "command": "node9",
      "args": ["mcp"]
    }
  }
}

FAQ

What is Node9?
Node9 is an execution security layer for AI agents that provides governance, monitoring, and audit logging. It sits between your AI agent and the tools it can use, allowing you to discover what it's been doing, protect against risky actions, and review activities over time.
How does Node9 work with MCP servers?
Node9 can wrap any MCP server transparently using its MCP gateway feature. The agent sees the same server, but Node9 intercepts every tool call to provide monitoring and governance. It also includes MCP tool pinning to prevent 'rug pull' attacks where servers might change their tool definitions unexpectedly.

On Hacker News

Recent discussion from the developer community.

Compare node9-proxy with

GitHub →

Last updated · Auto-generated from public README + GitHub signals.