timesfm
TimesFM (Time Series Foundation Model) is a pretrained time-series foundation model developed by Google Research for time-series forecasting.
Stealth web fetch, search and real Chrome control for AI agents.
教 AI 做这类工作的 Skill,可与本项一起使用
TimesFM (Time Series Foundation Model) is a pretrained time-series foundation model developed by Google Research for time-series forecasting.
Open-source intelligence for the global theater. Track everything from the corporate/private jets of the wealthy, and spy satellites, to seismic events in one unified interface. Hook an AI agent up to have it parse through data and find previously unseen correlations. The knowledge is available to…
scrape data from Google Maps. Extracts data such as the name, address, phone number, website URL, rating, reviews number, latitude and longitude, reviews,email and more for each place
列表按使用分(满分 100)排序,由以下部分组成
| 热度 | 9/80 | GitHub 星标或每周下载量:较大的计 60,较小的计 20 |
|---|---|---|
| 新鲜度 | 12/12 | 30 天内有发布或代码更新得满分 |
| 仍在响应 | 0/8 | 上次检查时云端端点有响应 |
| 使用分 | 21/100 |
质量: 基础检查通过 3/3 项(仅展示,不用于排序) · 上升势头: 暂无数据 (只用于“上升中”标签)
选择你使用的 AI 应用,然后复制设置。
用一句话让 AI 帮你安装
请帮我安装 WebSkrap。按照 https://mcp.popnx.com/i/kacigaya-webskrap.md 操作,先把命令给我看,等我同意后再运行。粘贴到能在你电脑上工作的 AI(如 Claude Code、Cursor)。它会读取我们的步骤,先给你看命令,等你同意。无需密钥,无需注册。 · 查看 AI 会读取的全文
通过 3/3 项检查 · 在你的电脑上运行 · ✓ 命令已检查 2026年11月17日
第一步:你用哪个 AI?
选择你用的应用,步骤会显示在下方。
可连接 Claude,但需要先准备
需要先在电脑上安装 Python (uv)(只需一次) · 仅限电脑上的 Claude 应用(Windows 或 Mac),网页和手机不可用
把这段话复制到 Claude 的对话里。AI 会先检查电脑,告诉你要改哪个文件,并且每次都等你同意。
请在这台电脑上为 Claude 安装 MCP server WebSkrap,按顺序:1) 用 uv --version 检查是否有 uv。没有的话,请让我从 https://docs.astral.sh/uv/getting-started/installation/ 安装,不要自己安装。2) 只使用这份配置(已通过 PopNx MCP 检查):{ "mcpServers": { "webskrap": { "command": "uvx", "args": [ "webskrap", "mcp" ] } } } 3) 告诉我要修改的文件完整路径或要运行的命令,等我同意。4) 不需要密钥。5) 按约定执行,然后确认 WebSkrap 的工具真的出现了。这段话不包含你的密钥。如果需要密钥,AI 会请你自己粘贴到文件或应用的输入框里。
已经装好了吗?先检查 · 打开终端,输入这条命令后按 Enter
uv --versionMac:按 ⌘ + Space,输入 Terminal 后按 Enter
Windows:打开“开始”菜单,输入 Windows PowerShell,然后选“打开”
还没安装: uv 没有点选式安装程序。从 uv 官方页面复制适合你系统(macOS 或 Windows)的一行命令,粘贴到终端 · 不需要单独安装 Python,uv 会在需要时自动下载
装好后关掉终端再打开,重新运行同一条命令,应显示版本号: uv --version
安装一次,所有需要 uv 的工具都能用
SettingsDeveloperEdit Config
{
"mcpServers": {
"webskrap": {
"command": "uvx",
"args": [
"webskrap",
"mcp"
]
}
}
}用 WebSkrap 告诉我你能帮我做什么,然后做一件给我看✓ 如果 Claude 说正在使用 WebSkrap,或回答中出现 WebSkrap 的工具名称,就表示连接成功
菜单名称已于 2026年11月11日 对照 Anthropic 帮助页面核实 · 你的界面可能因版本略有不同
ChatGPT 暂时无法连接 WebSkrap
WebSkrap 需要安装在你的电脑上,而 ChatGPT 只能连接在网上运行的服务器
A · 使用能连接 WebSkrap 的应用
B · 网页与浏览器 类中能连接 ChatGPT 的其他服务器
从同类中使用较多的里挑选,仅限用地址即可连接、无需密钥的
可连接 Claude Code,但需要先准备
需要先在电脑上安装 Python (uv)(只需一次)
把这段话复制到 Claude Code 的对话里。AI 会先检查电脑,告诉你要改哪个文件,并且每次都等你同意。
请在这台电脑上为 Claude Code 安装 MCP server WebSkrap,按顺序:1) 用 uv --version 检查是否有 uv。没有的话,请让我从 https://docs.astral.sh/uv/getting-started/installation/ 安装,不要自己安装。2) 只使用这份配置(已通过 PopNx MCP 检查):claude mcp add --scope user webskrap -- uvx webskrap mcp 3) 告诉我要修改的文件完整路径或要运行的命令,等我同意。4) 不需要密钥。5) 按约定执行,然后确认 WebSkrap 的工具真的出现了。这段话不包含你的密钥。如果需要密钥,AI 会请你自己粘贴到文件或应用的输入框里。
已经装好了吗?先检查 · 打开终端,输入这条命令后按 Enter
uv --versionMac:按 ⌘ + Space,输入 Terminal 后按 Enter
Windows:打开“开始”菜单,输入 Windows PowerShell,然后选“打开”
还没安装: uv 没有点选式安装程序。从 uv 官方页面复制适合你系统(macOS 或 Windows)的一行命令,粘贴到终端 · 不需要单独安装 Python,uv 会在需要时自动下载
装好后关掉终端再打开,重新运行同一条命令,应显示版本号: uv --version
安装一次,所有需要 uv 的工具都能用
Mac:按 Command+空格,输入 Terminal · Windows:打开 PowerShell
claude mcp add --scope user webskrap -- uvx webskrap mcp应能看到 WebSkrap 在列表中并显示已连接
用 WebSkrap 告诉我你能帮我做什么,然后做一件给我看✓ 如果 Claude Code 说正在使用 WebSkrap,或回答中出现 WebSkrap 的工具名称,就表示连接成功
菜单名称已于 2026年11月11日 对照 Anthropic 帮助页面核实 · 你的界面可能因版本略有不同
可连接 Cursor,但需要先准备
需要先在电脑上安装 Python (uv)(只需一次)
把这段话复制到 Cursor 的对话里。AI 会先检查电脑,告诉你要改哪个文件,并且每次都等你同意。
请在这台电脑上为 Cursor 安装 MCP server WebSkrap,按顺序:1) 用 uv --version 检查是否有 uv。没有的话,请让我从 https://docs.astral.sh/uv/getting-started/installation/ 安装,不要自己安装。2) 只使用这份配置(已通过 PopNx MCP 检查):{ "mcpServers": { "webskrap": { "command": "uvx", "args": [ "webskrap", "mcp" ] } } } 3) 告诉我要修改的文件完整路径或要运行的命令,等我同意。4) 不需要密钥。5) 按约定执行,然后确认 WebSkrap 的工具真的出现了。这段话不包含你的密钥。如果需要密钥,AI 会请你自己粘贴到文件或应用的输入框里。
已经装好了吗?先检查 · 打开终端,输入这条命令后按 Enter
uv --versionMac:按 ⌘ + Space,输入 Terminal 后按 Enter
Windows:打开“开始”菜单,输入 Windows PowerShell,然后选“打开”
还没安装: uv 没有点选式安装程序。从 uv 官方页面复制适合你系统(macOS 或 Windows)的一行命令,粘贴到终端 · 不需要单独安装 Python,uv 会在需要时自动下载
装好后关掉终端再打开,重新运行同一条命令,应显示版本号: uv --version
安装一次,所有需要 uv 的工具都能用
用 WebSkrap 告诉我你能帮我做什么,然后做一件给我看✓ 如果 Cursor 说正在使用 WebSkrap,或回答中出现 WebSkrap 的工具名称,就表示连接成功
菜单名称已于 2026年11月11日 对照 Cursor 帮助页面核实 · 你的界面可能因版本略有不同
可连接 VS Code,但需要先准备
需要先在电脑上安装 Python (uv)(只需一次)
把这段话复制到 VS Code 的对话里。AI 会先检查电脑,告诉你要改哪个文件,并且每次都等你同意。
请在这台电脑上为 VS Code 安装 MCP server WebSkrap,按顺序:1) 用 uv --version 检查是否有 uv。没有的话,请让我从 https://docs.astral.sh/uv/getting-started/installation/ 安装,不要自己安装。2) 只使用这份配置(已通过 PopNx MCP 检查):{ "servers": { "webskrap": { "type": "stdio", "command": "uvx", "args": [ "webskrap", "mcp" ] } } } 3) 告诉我要修改的文件完整路径或要运行的命令,等我同意。4) 不需要密钥。5) 按约定执行,然后确认 WebSkrap 的工具真的出现了。这段话不包含你的密钥。如果需要密钥,AI 会请你自己粘贴到文件或应用的输入框里。
已经装好了吗?先检查 · 打开终端,输入这条命令后按 Enter
uv --versionMac:按 ⌘ + Space,输入 Terminal 后按 Enter
Windows:打开“开始”菜单,输入 Windows PowerShell,然后选“打开”
还没安装: uv 没有点选式安装程序。从 uv 官方页面复制适合你系统(macOS 或 Windows)的一行命令,粘贴到终端 · 不需要单独安装 Python,uv 会在需要时自动下载
装好后关掉终端再打开,重新运行同一条命令,应显示版本号: uv --version
安装一次,所有需要 uv 的工具都能用
用 WebSkrap 告诉我你能帮我做什么,然后做一件给我看✓ 如果 VS Code 说正在使用 WebSkrap,或回答中出现 WebSkrap 的工具名称,就表示连接成功
菜单名称已于 2026年11月11日 对照 VS Code 帮助页面核实 · 你的界面可能因版本略有不同
Claude Code
claude mcp add --scope user webskrap -- uvx webskrap mcpClaude Desktop (claude_desktop_config.json)
{
"mcpServers": {
"webskrap": {
"command": "uvx",
"args": [
"webskrap",
"mcp"
]
}
}
}Cursor (~/.cursor/mcp.json)
{
"mcpServers": {
"webskrap": {
"command": "uvx",
"args": [
"webskrap",
"mcp"
]
}
}
}VS Code (mcp.json)
{
"servers": {
"webskrap": {
"type": "stdio",
"command": "uvx",
"args": [
"webskrap",
"mcp"
]
}
}
}这些设置是根据作者在注册中心的条目生成的。如果 README 的写法不同,请以 README 为准。
依据这个服务器的连接方式和各应用自己的文档,检查于 2026年11月10日。
在你的电脑上运行,适用于
ClaudeGooseLM StudioMsty StudioCherry StudioCursorVS Code + GitHub CopilotVisual StudioWindsurf (Devin Desktop)ZedKiroClineContinueJetBrains AI AssistantClaude CodeCodex CLIGemini CLIOpenCodenodecodeWarpLibreChatRaycast
它在你的电脑上运行;安装后,工具列表会显示在你的 AI 应用中。
我们检查的是任何人都能看到的信息;我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月19日
注册名称已通过 GitHub 账户 kacigaya 验证,该账户拥有此仓库
代码最后更新于昨天
使用 Apache-2.0
它在你的电脑上运行,没有互联网端点
在你的电脑上运行,改用本地设置
它在你的电脑上运行;我们只能读取互联网端点的工具列表
OSV.dev 中没有 webskrap@3.6.0 的记录
只检查这个版本的包本身,不包括所有依赖 · 检查于 2026年11月17日
如果你是作者,可把这段代码贴到 README。徽章会链接回本页。
[](https://mcp.popnx.com/en/servers/kacigaya-webskrap)提问、分享技巧或报告无法使用。按“有帮助”数量排序。
还没有评价
与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时,星标数才计入。
它在 3 项适用的基础检查中通过了 3 项。我们没有阅读代码。在添加密钥或令牌之前,请先看看这个服务器要求哪些权限,并使用只具备所需权限的密钥。
注册中心没有列出必需的密钥,不过有些服务器会在首次使用时要求你登录。
它在你的电脑上运行。你的 AI 应用会在需要时启动它,数据不会经过作者的服务器(除非它自己调用在线服务)。
GitHub 仓库最后更新于昨天,最新版本(3.6.0)于前天进入注册中心。
支持 MCP 的应用,例如 Claude Desktop、Claude Code、Cursor 和 VS Code。你的电脑上还需要有 Python (uv)。
同一分类中有:Browser Use, changedetection.io (official), Apify MCP Server, Pomerado, Actionbook
同一分类、其他作者、命令已通过检查。官方的排在前面,再按星标排序。
如何看卡片:↓ 每周下载量i★ GitHub 星标i三大支柱:评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i
Browser agents and stealth cloud browsers for AI: run web tasks, fill forms, extract data.
Official changedetection.io MCP: web page change tracking, back-in-stock and price alerts.
Extract data from any website with thousands of scrapers, crawlers, and automations on Apify Store ⚡
Connect your agents to any website. The cheaper, faster and more reliable alternative to browser use
Let your AI agent operate any website through a real, logged-in browser session.
| 本项 WebSkrap | Browser Use | changedetection.io (official) | Apify MCP Server | Pomerado | Actionbook | |
|---|---|---|---|---|---|---|
| 运行位置 | 需要安装在电脑上 | 云端或安装在电脑上都可以 | 在云端运行,无需安装 | 在云端运行,无需安装 | 在云端运行,无需安装 | 在云端运行,无需安装 |
| 需要密钥 | 否 | 是 | 否 | 是 | 否 | 否 |
| 可在 Claude 中按 + 添加 | 否 | 是 | 否 | 否 | 否 | 否 |
| 基础检查 | 3/3 | 5/5 | 3/3 | 4/4 | 4/4 | 4/4 |
| 许可证 | Apache-2.0 | MIT | Apache-2.0 | MIT | MIT | Apache-2.0 |
| 价格 | 登记中未列出费用 | 登记中未列出费用 | 登记中未列出费用 | 登记中未列出费用 | 登记中未列出费用 | 登记中未列出费用 |
按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。
WebSkrap
Async-first Python scraping on its own Chrome DevTools Protocol client, with persistent sessions, resource routing, stealth defaults, web search, a CLI, and an MCP server for agents.
pip install webskrap
webskrap install
webskrap install downloads Chrome for Testing. WebSkrap drives it through its own CDP client, with no Playwright. Upgrading from 2.x? Read the migration guide.
import asyncio
from webskrap import WebSkrapClient
async def main() -> None: async with WebSkrapClient() as client: result = await client.fetch("https://example.com") print(result.status, result.title) print(result.text[:200])
asyncio.run(main())
For deferred content, pass ready_selector="#content" to a Python fetch or --ready-selector '#content' to CLI fetch. Both MCP fetch tools accept ready_selector too. The selector must become visible after consent dismissal. Known challenge pages (DataDome, Cloudflare, HUMAN, Akamai, AWS WAF, and reCAPTCHA or hCaptcha on a 403, 429 or 503 page) and provider block pages (DataDome, Cloudflare, Akamai) raise blocked. WebSkrap waits for device checks that reload the page. It does not retry refusals. With solve_cloudflare=True (CLI --solve-cloudflare), it clicks Cloudflare's Turnstile checkbox; see Cloudflare Turnstile. With solve_datadome=True (CLI --solve-datadome), it drags DataDome's slider; see DataDome slider. With solve_perimeterx=True (CLI --solve-perimeterx), it holds HUMAN's Press & Hold button; see HUMAN Press & Hold. With solve_recaptcha=True (CLI --solve-recaptcha), it clicks a reCAPTCHA v2 or hCaptcha checkbox on a blocking page; see reCAPTCHA and hCaptcha checkbox. Ordinary HTTP errors still return ok=False. For an empty HTTP error body, Chromium can show its own error page. The returned text or HTML is that rendered page. The status, headers, and final_url still come from the HTTP response.
For any other CAPTCHA, complete it by hand: open a persistent browser with webskrap browser open --headed --session manual --proxy http://localhost:8888 URL. Complete the check in that browser. Keep the same session, profile, and proxy for later commands. On a remote host, use a desktop connection to its display. virtual_display=True alone does not provide a remote desktop.
Keep profile_mode="native" for the browser's own identity. On Linux with Xvfb, virtual_display=True runs a headed browser without a visible desktop window. Full profile emulation is useful for device tests, but cannot replace every worker or iframe value. See the identity limits.
Use client.fetch_many(urls) to fetch a finite batch in order with one browser:
async with WebSkrapClient() as client:
results = await client.fetch_many( ["https://example.com", "https://example.org"], text_only=True )
Each batch shares cookies and storage, then closes its browser. Separate batches and client.fetch() calls use fresh sessions. An explicit user_data_dir can still share state on disk. Empty batches do not launch a browser. The first page error stops the batch and raises; no partial list is returned. Ordinary HTTP errors remain results with ok=False. screenshot=True uses a separate generated PNG filename for each page.
For large inputs, per-page options, or state across batches, create a session with session = await client.session("crawl"). Call await session.fetch(url) for each URL. Close it with await client.close_session("crawl"), or let the client context close it. Session names belong to one client. Use separate sessions for unrelated callers.
Query fetched HTML with CSS selectors. Use parse_html(html) for HTML from other sources.
async with WebSkrapClient() as client:
result = await client.fetch("https://example.com") print(result.css_first("h1").text) print([a.attrs["href"] for a in result.css("a[href]")])
A search loads the engine's results page in the same stealth browser and returns the organic hits. Click-tracking redirects are unwrapped, so each url is the destination itself. Bing is the default; use engine="ddg" for DuckDuckGo's HTML endpoint. Google is not offered.
async with WebSkrapClient() as client:
result = await client.search("example domain", max_results=5) for hit in result.hits: print(hit.title, hit.url)
webskrap search "example domain" --engine ddg --max-results 10 --format json
When an engine distrusts the exit address it serves a bot challenge. That comes back as a blocked error, not as zero hits. Switch engine, reuse a persistent profile, or change the exit IP. WebSkrap does not retry.
solve_cloudflare=True (CLI --solve-cloudflare) handles Cloudflare's "Just a moment" page, like Scrapling's option of the same name. It is off by default; without it a Cloudflare challenge fails fast with blocked. With it, WebSkrap waits on the challenge page. When Cloudflare draws its Turnstile checkbox, the cursor wanders for 1.5 to 3 seconds, follows a curved path to the checkbox and clicks with a mouse-like button hold. The checkbox position is read from the widget's own DOM before each click. If the page has not cleared 20 seconds later, WebSkrap clicks again, at most three times. The checkbox gets up to 60 seconds from when it first appears, on top of timeout_ms. Success is reported only after the real page loads and no challenge is detected on it; otherwise the call fails with blocked.
It works in headless, headed and virtual_display modes. A click only answers the widget: Cloudflare still scores the browser and the exit IP, so a headless browser or a flagged IP can still be refused. No token is read, copied or injected, and no outside solving service is used. Other CAPTCHAs, such as image and audio puzzles, are reported as blocked and never touched.
solve_datadome=True (CLI --solve-datadome) handles DataDome's slider CAPTCHA. It is off by default; without it a DataDome CAPTCHA fails fast with blocked. With it, WebSkrap waits on the CAPTCHA page. When DataDome draws its slider, the cursor wanders for 3 to 5 seconds, moves to the handle, grabs it and drags it onto the slot in 35 small moves: slow, then faster, then slow again, with a little jitter. The handle and slot positions are read from the CAPTCHA frame's DOM before each drag. If the page has not cleared 10 seconds later, WebSkrap drags again, at most three times. The slider gets up to 60 seconds from when it first appears, on top of timeout_ms. Success is reported only after the real page loads and no challenge is detected on it; otherwise the call fails with blocked.
A drag only answers the slider: DataDome still scores the browser, the exit IP and the gesture. A hard block has no slider and stays blocked. Headed and virtual_display browsers passed in live runs; a headless browser can get the hard block instead. No cookie or token is read, copied or injected, no outside solving service is used, and DataDome's audio and image puzzles are never touched.
solve_perimeterx=True (CLI --solve-perimeterx) handles the Press & Hold challenge of HUMAN Security (formerly PerimeterX). It is off by default; without it the challenge page fails fast with blocked. With it, WebSkrap waits on the challenge page. When HUMAN draws its button, the cursor follows a curved path to it, presses, and holds with a small hand tremor until the button's progress fill is complete, then lets go. The button is read from the challenge frame's DOM before each hold. If the page has not cleared 8 seconds later, WebSkrap holds again, at most three times. The button gets up to 60 seconds from when it first appears, on top of timeout_ms. Success is reported only after the real page loads and no challenge is detected on it; otherwise the call fails with blocked.
A hold only answers the widget: HUMAN still scores the browser and the exit IP. In live runs it refused every completed hold from a flagged IP, in every browser mode, and accepted the first hold from a cleaner exit. No cookie or token is read, copied or injected, and no outside solving service is used.
solve_recaptcha=True (CLI --solve-recaptcha) clicks the "I'm not a robot" checkbox of reCAPTCHA v2 or hCaptcha on a page that blocks the site: a 403, 429 or 503 response that carries the widget. It is off by default; without it such a page fails fast with blocked. With it, WebSkrap waits for the widget, moves the cursor like a person and clicks the checkbox. When the provider trusts the browser, the box ticks and the page's own callback continues to the real page. If the page has not cleared 15 seconds later, WebSkrap clicks again, at most twice within 60 seconds. When the provider opens an image challenge instead, WebSkrap leaves it alone and the call fails with blocked at once.
For a checkbox inside a form, fill the form, then call browser_solve_captcha (CLI webskrap browser solve-captcha) before you submit it. Its captcha.state is checked, challenge, gone, timeout or not_found.
The click only answers the checkbox: the provider still scores the browser and the exit IP, and a datacenter IP usually gets the image challenge. reCAPTCHA v3 and invisible widgets have no checkbox and are not handled. No token is read, copied or injected, and no outside solving service is used.
Use them responsibly: only on sites whose terms and applicable law allow automated access, and at a polite rate. Passing a challenge does not grant permission to collect data you may not collect.
webskrap fetch https://example.com --virtual-display --solve-cloudflare
webskrap fetch https://example.com --virtual-display --solve-datadome webskrap fetch https://example.com --virtual-display --solve-perimeterx webskrap fetch https://example.com --virtual-display --solve-recaptcha webskrap browser solve-captcha
Agents can see the rendered viewport and interact by coordinates, including canvas and controls missing from the accessibility tree. MCP browser_view returns a PNG image directly; browser_mouse clicks, moves, scrolls, or drags. browser_insert_text inserts literal text into the focused control.
webskrap browser open https://example.com
webskrap browser view viewport.png --format json
webskrap browser mouse click 120 80 webskrap browser insert-text "hello" webskrap browser press