mcp-server-webcrawl

Name: mcp-server-webcrawl
Author: pragmar

@pragmar

代码仓库主页

关于 mcp-server-webcrawl

MCP server tailored to connecting web crawler data and archives

#web-crawler #web-scraping

基本信息

分类

其他

许可证

NOASSERTION

运行时

python

传输方式

stdio

发布者

pragmar

提交者

pragmar

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "webcrawl": {
      "command": "mcp-server-webcrawl",
      "args": [
        "--crawler",
        "wget",
        "--datasrc",
        "/path/to/wget/archives/"
      ]
    }
  }
}

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is mcp-server-webcrawl?

mcp-server-webcrawl provides advanced search and retrieval for web crawler data, enabling an AI client to filter and analyze web content. It includes a fulltext search interface with boolean support and resource filtering by type, HTTP status, and more. It works with various web crawlers such as ArchiveBox, HTTrack, InterroBot, Katana, SiteOne, WARC, and wget.

How to use mcp-server-webcrawl?

Install via pip: pip install mcp-server-webcrawl. Requires Python (>=3.10) and Claude Desktop. For step-by-step MCP server setup, refer to the Setup Guides. The server can also be used in interactive terminal mode for classic boolean search without AI.

Key features of mcp-server-webcrawl

Multi-crawler compatible (ArchiveBox, HTTrack, InterroBot, Katana, SiteOne, WARC, wget)
Boolean search support with field-specific queries
Filter by type, status, and more
Extras: markdown, snippets, regex, xpath, thumbnails
Prompt routines provided (SEO audit, 404 audit, Gopher interface, etc.)
Interactive mode for terminal-based searching of web archives

Use cases of mcp-server-webcrawl

Analyze crawled website content with fulltext and boolean queries
Perform technical SEO audits using provided prompt routines
Detect broken links and suggest fixes with 404 audit
Build a roll-your-own website knowledgebase from crawled data
Search web archives remotely via SSH interactive mode

FAQ from mcp-server-webcrawl

What crawlers are supported?

mcp-server-webcrawl works with ArchiveBox, HTTrack, InterroBot, Katana, SiteOne, WARC, and wget. Each has a dedicated setup guide.

What are the runtime requirements?

The server requires Python >=3.10 and Claude Desktop. It is installed via pip install mcp-server-webcrawl.

Can I use it without an AI?

Yes, mcp-server-webcrawl includes an interactive mode that runs classic Boolean search on your web archives in a terminal, no AI required.

What extras help reduce token usage?

Markdown, snippets, regex, xpath, and thumbnails can transform HTTP data to be more token-efficient. For example, markdown produces about one-third the bytes of source HTML.

How does boolean search work?

The query engine supports field-specific searches (e.g., type: html, status: 200) and boolean operators (AND, OR, NOT). Fulltext can combine url, content, and headers fields. Wildcards and exact phrases are also supported.