Site MCP
Turn any public website into an MCP server for agents to search, read and navigate.
我该使用它吗
质量与安全性
发现（8）
- HIGHTool poisoning patterns detected
- MEDIUMTool description contains URL to non-standard domain在 list_pages 中
- MEDIUMTool description contains URL to non-standard domain在 read_page 中
- MEDIUMTool description contains URL to non-standard domain在 search_site 中
- MEDIUMTool description contains URL to non-standard domain在 get_site_info 中
- MEDIUMTool description contains URL to non-standard domain在 get_page_info 中
- MEDIUMTool description contains URL to non-standard domain在 list_links 中
- LOWTool 'list_links' description lacks action verb在 list_links 中
基于对工具定义和协议合规性的自动分析。
上下文开销
这是每次将服务器的工具加载到模型上下文窗口时所消耗的大致 token 数。数值越高，可用于其他任务的注意力就越少。
安装
一键安装
将以下内容添加到你的 `claude_desktop_config.json` 文件中：
{
"mcpServers": {
"site-mcp": {
"url": "https://sitemcp.dev/sitemcp.dev"
}
}
}远程端点
https://sitemcp.dev/sitemcp.devstreamable-http它能做什么
工具清单
工具（6）
🟢list_pages(filter, limit)
List page URLs on https://sitemcp.dev/ (from its sitemap, or links on the homepage).
输入模式
{
"type": "object",
"properties": {
"filter": {
"description": "Only return URLs containing this substring",
"type": "string"
},
"limit": {
"default": 100,
"type": "integer",
"minimum": 1,
"maximum": 500
}
},
"$schema": "http://json-schema.org/draft-07/schema#"
}🟢read_page(path, offset)
Fetch a page from https://sitemcp.dev/ live and return its main content as Markdown, with when it was read and when the page says it last changed. Long pages are paginated with offset.
输入模式
{
"type": "object",
"properties": {
"path": {
"default": "/",
"description": "Path like \"/pricing\" or a full URL on sitemcp.dev",
"type": "string"
},
"offset": {
"default": 0,
"description": "Character offset for long pages",
"type": "integer",
"minimum": 0,
"maximum": 9007199254740991
}
},
"$schema": "http://json-schema.org/draft-07/schema#"
}🟢search_site(query, max_results)
Search https://sitemcp.dev/ for pages relevant to a query. Returns matching pages with snippets, each with when it was read and when the page says it last changed.
输入模式
{
"type": "object",
"properties": {
"query": {
"type": "string",
"minLength": 1
},
"max_results": {
"default": 5,
"type": "integer",
"minimum": 1,
"maximum": 10
}
},
"required": [
"query"
],
"$schema": "http://json-schema.org/draft-07/schema#"
}🟢get_site_info
Orientation for https://sitemcp.dev/: title, description, language, sitemaps, robots.txt, pages discovered and search-index status. Call this first.
输入模式
{
"type": "object",
"properties": {},
"$schema": "http://json-schema.org/draft-07/schema#"
}🟢get_page_info(path)
Metadata for one page on https://sitemcp.dev/ without its full text: title, description, headings, canonical URL, published and modified dates, content type, language and word count. Use it to decide whether a page is worth reading.
输入模式
{
"type": "object",
"properties": {
"path": {
"default": "/",
"description": "Path like \"/pricing\" or a full URL on sitemcp.dev",
"type": "string"
}
},
"$schema": "http://json-schema.org/draft-07/schema#"
}🟢list_links(path, scope)
Every link on one page of https://sitemcp.dev/, with its text and whether it stays on the site. Use it to navigate the site instead of guessing URLs.
输入模式
{
"type": "object",
"properties": {
"path": {
"default": "/",
"description": "Path like \"/docs\" or a full URL on sitemcp.dev",
"type": "string"
},
"scope": {
"default": "all",
"type": "string",
"enum": [
"all",
"internal",
"external"
]
}
},
"$schema": "http://json-schema.org/draft-07/schema#"
}社区
证据