# Check Point Management Write-Path MCP Server

> Draft/publish/install access-rule lifecycle for compensating-control workflows -- the write path Check Point's official read-only MCP does not expose.

- Type: MCP server
- Author: tarhou
- License: MIT
- Trust tier: contributed (passed automated validation and a baseline maintainer review)
- Last reviewed: 2026-07-31
- Source repository: https://github.com/tarhou/checkpoint-mcp-server
- Repo state: not archived
- Last push: 2026-07-29T08:02:46Z
- Stars: 0 (a popularity signal, not a safety signal)
- Forks: 0 (a popularity signal, not a safety signal)
- Open issues: 0
- Language: Python
- Health checked at: 2026-11-05T20:35:15.821Z
- Integrations: Check Point
- Domains: network-security
- Tags: firewall, compensating-controls, check-point, network-security, python
- Date added: 2026-07-29
- Transport: stdio
- Runtime: python
- Auth method: none
- Compatible clients: Claude Code, Claude Desktop
- Tools:
  - `checkpoint_list_access_rules`: List current rules in an access layer
  - `checkpoint_add_access_rule`: Add a rule in the current session (status: draft until published)
  - `checkpoint_publish_session`: Commit drafted rules to the management server (draft to published)
  - `checkpoint_install_policy`: Push published rules to gateways (published to installed -- the rule actually enforces)
- Listing page: https://exchange.tenable.com/mcp-servers/checkpoint-mcp-server
- Source file: https://github.com/tenable/cyberagents-exchange/blob/main/mcp-servers/checkpoint-mcp-server.md

An MCP server for **Check Point Management** write-path workflows: draft an access rule, publish the session, install policy — the full compensating-control lifecycle as typed MCP tools. Originally built for a live orchestration demo at Tenable EXPOSURE 2026 (Boston), where an AI agent deployed a compensating firewall control for an unpatchable industrial asset under human supervision.

## What it does

Check Point's official MCP bundle (`@chkp/quantum-management-mcp`) is read-only by design. But the classic vulnerability-management move for an asset you cannot patch — block traffic to it — needs `add-access-rule`, `publish`, and `install-policy`. This server fills that write-side gap with the three-stage lifecycle (draft → published → installed) matching the real Management API state machine, so an AI agent's audit trail reads exactly like a human operator's, and a supervising human gets two natural gates before anything enforces.

## How it works

FastMCP v3 server whose tool surface mirrors Check Point's Management API (`add-access-rule`, `publish`, `install-policy`, `show-access-rulebase`). The backend is deliberately in-memory — safe for demos, agent development, and workflow testing with zero credentials — and the interfaces are designed so the official `cp_mgmt_api_python_sdk` slots in for production against Smart-1 Cloud or on-prem without changing tool signatures or response shapes. All outputs are Markdown rule tables and lifecycle status summaries.

This is the Markdown version of one of 140 listings on CyberAgents Exchange. The same listing as JSON: https://exchange.tenable.com/api/mcp-servers/checkpoint-mcp-server.json. The full index and how to read it: https://exchange.tenable.com/llms.txt