网页抓取 MCP 推荐

轻松跟踪与浏览 RSS 订阅。从任意订阅地址获取最新条目并提取完整…文本。

非官方的DeepWiki MCP服务器，通过MCP接收DeepWiki URL，爬取所有相关页面并转换为Markdown格式，可返回单个文档或按页面列表返回

中文站能否被豆包/DeepSeek/文心/Kimi 抓取、读懂、引用：可引用性审计、信源池占位探测、五大指标测量、修复计划、观测时序追踪

pubmed-mcp-server（文献检索）

检索 PubMed/Europe PMC，获取文章与全文（PMC/EPMC/Unpaywall）、引文与 MeSH 主题词。

paper-search-mcp-openai-v2

从 arXiv、PubMed、bioRxiv、medRxiv、Google Scholar、Semantic Scholar、CrossRef 和 IACR 等主流来源查找并下载学术论文。返回标准化结果，并在可用时获取 PDF 全文。以深度检索与便捷获取加速文献综述。

arxiv-mcp-server

检索 arXiv 论文，获取论文元数据并读取全文内容。

Screenshot & PDF Capture — Full-Page Chromium Rendering

Web capture API for AI agents. Take full-page screenshots (PNG/JPEG/WebP) and generate PDFs from any URL using headless Chromium. Custom viewport size, device emulation, and scroll capture. Tools: capture_screenshot (image), webpage_to_pdf (PDF). Use this for visual regression testing, documentation, archiving web pages, or generating reports. IMPORTANT: For text extraction from pages, use web_scrape_to_markdown. For custom PDF generation from data, use document_generate_pdf. Returns: base64-encoded image or PDF. No API key required — x402 micropayment $0.008/call on Base L2.

youtube-mcp

Search and browse videos, channels, and playlists to fetch titles, descriptions, stats, and durations. Retrieve multilingual, timestamped transcripts and search within captions for precise context. Surface channel and playlist insights quickly by listing items and details.

Web Auditor 网站审计

分析单个网页乃至整站在 SEO、可访问性、性能与安全合规方面的表现。生成详细报告与可执行的问题清单，改善站点质量与用户体验。可跟踪历史审计数据，在多次抓取之间监测改进情况。

WorldMonitor 全球情报

以 74 个工具的 MCP 服务器提供实时全球情报：市场行情、冲突事件、国家风险与韧性评分、航运咽喉要道、海事、航空与机票定价、能源、气候、健康、网络威胁、制裁、贸易流向、预测，以及 AI 生成的全球/国别简报。另附 25 个 Agent Skills、6 个工作流提示、11 个资源与 10 个交互式聊天内应用。发现过程匿名——无需密钥即可连接并调用 tools/list、prompts/list、skills/list。几乎所有工具都支持 JMESPath 投影，智能体只取所需字段。get_sources 免费匿名；登录账号每天可少量使用缓存类工具；完整访问需 OAuth2 或从 worldmonitor.app/pro 获取 API 密钥。

Meta-Stamp Pockets 授权内容库

面向 AI 智能体的授权创作者内容。以预先索引的 Pocket 形式发现音乐、视频与文本目录——只需数百 token 而非上万 token 的抓取，30 毫秒内返回。发现功能免费；内容拉取通过 HTTP 402 付费墙按次授权——连接即合约，每次拉取收入的 85% 即时结算给版权方。工具：list_pockets、search_pockets、pull_content。浏览无需密钥；拉取内容会返回带注册说明的 402 响应——按提供的 URL 注册、以 Bearer token 附上 API 密钥，即可拉取带完整署名与审计链路的授权内容。

crossref-mcp-server（学术索引）

解析 DOI，通过 Crossref REST API 检索约 1.55 亿条学术作品并获取参考文献。

Simplify common data manipulation tasks like encoding, hashing, and formatting across various formats. Convert between CSV, JSON, Markdown, and HTML seamlessly to streamline data workflows. Extract insights from text and configurations through robust parsing, regex testing, and statistical analysis.

ToolSnap MCP

Context-efficient microtools for AI agents. 33 tools total. Flagship fetch_extract: 98.1% median token reduction (53,820 → 2,001 tokens) — saves ~$0.156/call at Sonnet pricing vs loading raw HTML. Paid tools: web extraction, full-page screenshots, background removal (fal.ai rembg), keyword research (DataForSEO), PDF text extract, HTML-to-Markdown, page asset/link inventory, temporary file upload. 20 always-free utility tools: UUID, hash, Base64, URL encode/decode, JSON format/query, timestamps, text stats, regex extract, CSV/JSON query, RSS/sitemap parse, and more. Pricing: $0.02 USDC on Base via x402 (EIP-3009), or $0.01 prepaid (deposit ≥$0.50 USDC, off-chain debit, no per-call gas). First call free per wallet on most tools. Includes pay-proxy: local stdio bridge that auto-signs x402 challenges — any MCP client gains payment capability without code changes or API keys.

openlibrary-mcp-server（开放图书馆）

检索图书与作者、获取版本信息、浏览主题并解析封面图片。

Compress.new

Convert any webpage to clean markdown and feed it directly into AI agent workflows. Why This Matters? Adding webpages to LLM conversations usually means dumping raw HTML, bloated with ads, scripts, and formatting noise. This MCP integrates compress.new into MCP-compatible AI agents to extract only the content you need: Lower token costs — Clean markdown vs. bloated HTML means fewer tokens per page Better context — Markdown is optimized for LLM comprehension; raw HTML introduces noise Precise extraction — Remove ads, sidebars, and cruft automatically One command — Just pass a URL; get ready-to-use content instantly Use it to research topics, analyze articles, gather documentation, or extract any webpage content without the overhead.

邮件发送器 —— 通过 API 发交易型邮件

面向 AI 智能体的邮件发送 API。以编程方式发送交易型邮件——支持文本或 HTML 正文、自定义主题与发件地址，并返回投递状态确认与消息 ID，由 Resend 提供能力。工具：email_send_message。当智能体需要发送通知、报告、告警或外呼邮件时使用。注意：发送前先用 email_verify_address 校验收件地址以避免退信。返回：{id, status, from, to, subject}。无需 API 密钥——在 Base L2 上以 x402 微支付 $0.003/次。

提供 smartfetch 直链抓取、smartsearch 网络发现、smartcrawl 站点遍历与文档导出的 MCP。

md-log 的 MCP 服务：按路径保存、更新、获取 Markdown 文档（stdio 与远程 HTTP，PAT 鉴权）。

OrchestKit 文档的只读 MCP 服务器：全文检索加 Markdown 原文获取，无需鉴权。

Cron jobs with retries, a webhook relay with replay, and HTML-to-PDF with hosted templates.

只读、经核实的可被代理操作的 GTM 工具记录：搜索、抓取、对比并跟踪变更。

通过 MCP 访问大都会艺术博物馆馆藏——50 万件以上艺术品、元数据、来源与开放获取图像。

检索 PubMed/Europe PMC，获取文章与全文（PMC/EPMC/Unpaywall）、引文与 MeSH 主题词。

Markdown editor in one offline HTML file: the agent edits passages, you keep or drop each change.

跨 17 个数据档案库、组学注册库与文献源查找并获取科研数据集。

通过 MCP 从 FRED API 检索约 80 万条美联储经济时间序列。

MCP server for Google Workspace — Sheets, Drive, Docs (HTML or Markdown), and Calendar.

Publish HTML reports to a share link, collect anchored comments, revise at the same URL.

128 个 REST 操作。120 条 MCP 路由；119 个 JSON/文本操作。OAuth 2.1。与 X Corp 无关联。

Public web data for AI agents: scrape any page, plus Google, TikTok, Instagram and Amazon as JSON.

Live web access for agents: scrape, SERP search, crawl/map, 100+ collectors, datasets, proxies.

Turn a cover and a short animation, supplied as HTML, into an Apple Live Photo, an MP4 and a GIF.

Create contact forms, get HTML/React/Next.js snippets and read recent submissions on Formgong.

为 Claude 提供 Etix 活动发现能力：检索活动、场馆与演出者，并获取活动详情。

为 Claude 提供只读的 Workday 访问：借助你已登录的会话获取待办、薪酬与福利信息。

Dark web OSINT over Tor: search onion engines, scrape pages, report with your own model.

Any file → clean Markdown for AI agents: PDF, Office, EPUB, HTML, images, audio/video. Local MCP.

Publish HTML pages, collect review comments, and revise at the same link.

Read any page past its anti-bot wall, crawl sites, search without an API key, solve captchas locally

把 Markdown 渲染为适合邮件的 HTML，检查草稿中影响送达率的问题，并预览邮件效果。

Search CERN Open Data, fetch records, files, analysis environments, CMS good-run lists, HLT paths.

AI website & landing page builder: animated site in an 8B design, preview, edits, HTML download

Scan URLs or HTML for WCAG 2.2 violations. 75-rule manifest, weighted score, shareable reports.

搜索 plati.market（Digiseller）商品目录，并获取含卖家信息的商品详情。

面向 AI 智能体的设计风格库：检索真实风格，获取可直接套用的设计规范。

Web scraping, SERP search, crawl/map and 74 Collectors through residential proxies. 25 tools.

Publish HTML, Markdown, decks and dashboards as private pages; share them with named people.