Skip to main content
Glama

PrismSec 🔷

安全、模块化的 MCP 服务器,用于渗透测试工具。

将 7 款行业标准安全工具(nmap、nuclei、gobuster、subfinder、httpx、nikto、sqlmap)封装为 13 个已注册的 MCP 工具——可直接与 Claude、Cursor、Copilot 及任何兼容 MCP 的 AI 智能体一起使用。


功能

功能

描述

零 shell=True

所有子进程调用均使用 asyncio.create_subprocess_exec——无 shell 注入风险

输入验证

目标、URL、端口、严重级别——执行前均经过验证

注入检测

阻止 shell 元字符(;$()、反引号、|

超时强制执行

每个工具都有可配置的超时时间——自动终止挂起的进程

结构化输出

将 XML/JSON/文本解析为干净的 JSON,供 AI 智能体使用

模块化架构

每个工具一个文件——易于添加、维护和测试

MCP SDK v2

基于最新的 Model Context Protocol SDK 构建


Related MCP server: Kali Tools MCP Server

安装

从源码安装

git clone https://github.com/azmisyahrul/prismsec.git
cd prismsec
pip install -e .

前提条件

安装你所需的安全工具:

# Ubuntu/Debian
apt install nmap nikto sqlmap

# Go-based tools
go install -v github.com/projectdiscovery/nuclei/v3/cmd/nuclei@latest
go install -v github.com/projectdiscovery/subfinder/v2/cmd/subfinder@latest
go install -v github.com/projectdiscovery/httpx/cmd/httpx@latest
go install github.com/OJ/gobuster/v3@latest

工具(13 个已注册)

Nmap — 端口扫描

MCP 工具

描述

nmap_port_scan

使用快速/完整/服务/隐蔽/激进模式进行端口扫描

nmap_service_detect

检测开放端口的服务/版本

nmap_full_scan

扫描全部 65535 个 TCP 端口

Nuclei — 漏洞扫描

MCP 工具

描述

nuclei_vuln_scan

使用所有模板进行全面漏洞扫描

nuclei_severity_scan

按严重级别过滤的扫描(critical、high 等)

nuclei_template_scan

使用特定模板进行定向扫描

Gobuster — 目录/DNS 爆破

MCP 工具

描述

gobuster_directory

使用可配置的扩展名进行目录暴力破解

gobuster_dns

DNS 子域暴力破解

其他工具

MCP 工具

描述

subfinder_enumerate

被动子域枚举(crt.sh、VirusTotal 等)

httpx_probe

Web 探测——存活检测、标题、技术指纹识别

nikto_web_scan

Web 服务器漏洞扫描

sqlmap_injection_test

SQL 注入检测与测试

元信息

MCP 工具

描述

check_tools

检查已安装了哪些安全工具


用法

Claude Desktop

添加到 claude_desktop_config.json

{
  "mcpServers": {
    "prismsec": {
      "command": "python3",
      "args": ["/path/to/prismsec/server.py"],
      "env": {}
    }
  }
}

Claude Code

claude mcp add prismsec python3 /path/to/prismsec/server.py

Cursor / Windsurf / Cline

添加到 .cursor/mcp.json 或等效配置:

{
  "mcpServers": {
    "prismsec": {
      "command": "python3",
      "args": ["/path/to/prismsec/server.py"]
    }
  }
}

SSE 传输(远程)

# Server side
python3 server.py --transport sse --host 0.0.0.0 --port 8000

# Client config
{
  "mcpServers": {
    "prismsec": {
      "url": "http://localhost:8000/sse"
    }
  }
}

项目结构

prismsec/
├── server.py              # MCP server entry point (13 tools)
├── pyproject.toml         # Project config + dependencies
├── tools/                 # Tool wrappers (one file per tool)
│   ├── base.py           # ToolWrapper ABC + async runner
│   ├── nmap.py           # Nmap — XML parsing, scan modes
│   ├── nuclei.py         # Nuclei — JSON output parsing
│   ├── gobuster.py       # Gobuster — text output parsing
│   ├── subfinder.py      # Subfinder — subdomain enum
│   ├── httpx.py          # Httpx — web probing
│   ├── nikto.py          # Nikto — web vuln scan
│   └── sqlmap.py         # Sqlmap — SQL injection testing
├── parsers/               # Output parsers
│   ├── xml_parser.py     # nmap XML → structured JSON
│   ├── json_parser.py    # JSON/JSONL parsing
│   └── text_parser.py    # Gobuster, nikto, sqlmap text
└── utils/                 # Shared utilities
    ├── runner.py          # AsyncRunner with timeout
    ├── validator.py       # Input validation + injection detection
    ├── rate_limiter.py    # Token bucket rate limiter
    └── logging.py         # Structured logging

配置

环境变量

默认值

描述

LOG_LEVEL

INFO

日志级别(DEBUG、INFO、WARNING、ERROR)


安全注意事项

⚠️ 仅限授权测试。 仅对你自己拥有或已获得书面许可的系统使用。

  • 工具输出可能包含敏感信息(IP 地址、开放端口、漏洞)

  • 服务器默认绑定到 127.0.0.1——切勿暴露给不受信任的网络

  • 每个工具都有可配置的超时时间,以防止资源耗尽


许可证

MIT


基于 Model Context Protocol 标准构建,以实现广泛的客户端兼容性。

A
license - permissive license
Not graded
quality - not tested
C
maintenance

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Servers

  • F
    license
    Not graded
    quality
    B
    maintenance
    An MCP server that exposes over 20 standard penetration testing utilities, such as Nmap, SQLMap, and OWASP ZAP, as callable tools for AI agents. It enables natural language control over complex security workflows for automated and interactive penetration testing.
    91
  • A
    license
    Not graded
    quality
    D
    maintenance
    Integrates 7 security tools (nmap, nuclei, dirsearch, sqlmap, hydra, Acunetix, Metasploit) via MCP protocol for AI-assisted penetration testing with enterprise-grade safety features.
    1
    MIT
  • F
    license
    Not graded
    quality
    D
    maintenance
    AI-powered Attack Surface Intelligence server that exposes industry-standard penetration testing tools via MCP, enabling AI agents to perform comprehensive security assessments.
    3
  • A
    license
    C
    quality
    C
    maintenance
    AI-powered security scanning MCP server that exposes 25+ professional tools, enabling penetration testing and security assessments through natural language interaction with AI agents like Claude Desktop.
    31
    MIT

View all related MCP servers

Related MCP Connectors

  • Security scanner for MCP servers. Detect vulnerabilities, prompt injection, and tool poisoning.

  • Scans MCP servers for tool poisoning, prompt injection and supply chain risks.

  • MCP server for Pentest-Tools.com: run scans, manage findings and reports via your preffered LLM.

View all MCP Connectors

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/azmisyahrul/prismsec'

If you have feedback or need assistance with the MCP directory API, please join our Discord server