Web Check MCP
MCP server that exposes 31 OSINT checks from Lissy93/web-check as tools for website analysis, including SSL, DNS, headers, WHOIS, and security presets. Enables natural-language-driven web recon and health checks.
README
Web Check MCP
<!-- mcp-name: io.github.AMEOBIUS-space/web-check-mcp -->
Agent wrapper for Lissy93/web-check — 31 OSINT checks as MCP tools + CLI.
Gap verified 2026-07-25: no public MCP/SDK/Hermes skill existed for this API. This package fills it.
Features
- 8 MCP tools:
webcheck_run,webcheck_ssl,webcheck_dns,webcheck_security,webcheck_headers,webcheck_whois,webcheck_list_checks,webcheck_health - 31 OpenAPI checks from upstream (
ssl,dns,headers,ports,firewall,whois, …) - Presets:
quick|security|server|quality|heavy|all - Parallel fan-out + payload truncation (agent-context safe)
- Zero pip deps — Python 3.10+ stdlib only
- STDIO JSON-RPC with
initializehandshake + tool annotations (readOnlyHint+openWorldHint)
Public API works (with fallback)
Two public bases are tried automatically:
| Base | Status |
|---|---|
web-check.as93.net/api (Netlify mirror) |
200 OK — more open, Cloudflare front |
web-check.xyz/api (Vercel primary) |
200/429 — may challenge datacenter IPs |
Set WEB_CHECK_BASE_URL explicitly to skip probing. Fallback auto-enables when base starts with https://web-check. and can be forced via WebCheckClient(fallback=True).
No key, no auth.
Self-host (optional)
For heavy load or offline use:
docker run -d --name web-check -p 3000:3000 lissy93/web-check
export WEB_CHECK_BASE_URL=http://127.0.0.1:3000/api
Or compose (upstream + thin MCP image):
docker compose up -d web-check
# MCP stdio still launched by host; see Dockerfile ENTRYPOINT
Upstream Go rewrite xray-web/web-check-api is early WIP — prefer Node/Docker image for full endpoint parity.
Quick Start
# from repo root
pip install -e .
# or: python3 -m pip install web-check-mcp
# Manifest
python3 -m src.server --manifest
# List checks
python3 -m src.server list --group quick
# Health probe
python3 -m src.server health
# Run quick recon (needs live API)
python3 -m src.server run example.com --group quick
# Single check
python3 -m src.server check ssl example.com
# MCP STDIO (Claude Desktop / Hermes / Cursor)
python3 -m src.server --stdio
Library
from src.client import WebCheckClient
client = WebCheckClient(base_url="http://127.0.0.1:3000/api")
print(client.run("example.com", group="quick"))
print(client.check_one("ssl", "example.com"))
MCP client config
{
"mcpServers": {
"web-check": {
"command": "python3",
"args": ["-m", "src.server", "--stdio"],
"cwd": "/absolute/path/to/web-check-mcp",
"env": {
"WEB_CHECK_BASE_URL": "http://127.0.0.1:3000/api"
}
}
}
}
After pip install web-check-mcp:
{
"mcpServers": {
"web-check": {
"command": "web-check-mcp",
"args": ["--stdio"],
"env": {
"WEB_CHECK_BASE_URL": "https://web-check.as93.net/api"
}
}
}
}
Official registry metadata lives in server.json (io.github.AMEOBIUS-space/web-check-mcp). Publish with mcp-publisher after the next PyPI release that includes the mcp-name marker above.
Tool Reference
| Tool | Description |
|---|---|
webcheck_list_checks |
Catalog of endpoints / groups |
webcheck_health |
Probe API base reachability |
webcheck_run |
Parallel multi-check (default group=quick) |
webcheck_ssl |
Certificate chain |
webcheck_dns |
DNS records |
webcheck_security |
Security preset bundle |
webcheck_headers |
HTTP headers |
webcheck_whois |
Domain WHOIS |
Env: WEB_CHECK_BASE_URL, WEB_CHECK_TIMEOUT, WEB_CHECK_MAX_WORKERS, WEB_CHECK_MAX_CHARS.
Tests
python3 -m pytest tests/ -v
All network paths mocked — no live API required for CI.
License
MIT. Upstream Web Check © Alicia Sykes, MIT.
Not affiliated with Lissy93; thin agent-facing client only.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。