timesfm
TimesFM (Time Series Foundation Model) is a pretrained time-series foundation model developed by Google Research for time-series forecasting.
Browser MCP for the Chrome you already have open, via the free Qiksy Bridge extension.
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 | 12/80 | 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 | 24/100 |
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 Qiksy Bridge。按照 https://mcp.popnx.com/i/qiksy-bridge.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 Node.js(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 Qiksy Bridge,按顺序:1) 用 node -v 检查是否有 Node.js。没有的话,请让我从 https://nodejs.org/en/download 安装,不要自己安装。2) 只使用这份配置(已通过 PopNx MCP 检查):{ "mcpServers": { "bridge": { "command": "npx", "args": [ "-y", "qiksy-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 Qiksy Bridge 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
node -vMac: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: 打开官方下载页,选择 LTS 版本和点选式安装程序:Mac 用 .pkg · Windows 用 .msi。双击后一路下一步即可,不需要 curl 命令
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: node -v
安装一次,所有需要 Node.js 的工具都能用
SettingsDeveloperEdit Config
{
"mcpServers": {
"bridge": {
"command": "npx",
"args": [
"-y",
"qiksy-mcp"
]
}
}
}用 Qiksy Bridge 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 Qiksy Bridge,或回答中出现 Qiksy Bridge 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 Qiksy Bridge
Qiksy Bridge 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 Qiksy Bridge 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 Node.js(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 Qiksy Bridge,按顺序:1) 用 node -v 检查是否有 Node.js。没有的话,请让我从 https://nodejs.org/en/download 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 bridge -- npx -y qiksy-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 Qiksy Bridge 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
node -vMac: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: 打开官方下载页,选择 LTS 版本和点选式安装程序:Mac 用 .pkg · Windows 用 .msi。双击后一路下一步即可,不需要 curl 命令
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: node -v
安装一次,所有需要 Node.js 的工具都能用
Mac:按 Command+空格,输入 Terminal · Windows:打开 PowerShell
claude mcp add --scope user bridge -- npx -y qiksy-mcp应能看到 Qiksy Bridge 在列表中并显示已连接
用 Qiksy Bridge 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 Qiksy Bridge,或回答中出现 Qiksy Bridge 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 Node.js(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 Qiksy Bridge,按顺序:1) 用 node -v 检查是否有 Node.js。没有的话,请让我从 https://nodejs.org/en/download 安装,不要自己安装。2) 只使用这份配置(已通过 PopNx MCP 检查):{ "mcpServers": { "bridge": { "command": "npx", "args": [ "-y", "qiksy-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 Qiksy Bridge 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
node -vMac: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: 打开官方下载页,选择 LTS 版本和点选式安装程序:Mac 用 .pkg · Windows 用 .msi。双击后一路下一步即可,不需要 curl 命令
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: node -v
安装一次,所有需要 Node.js 的工具都能用
用 Qiksy Bridge 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 Qiksy Bridge,或回答中出现 Qiksy Bridge 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 Node.js(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 Qiksy Bridge,按顺序:1) 用 node -v 检查是否有 Node.js。没有的话,请让我从 https://nodejs.org/en/download 安装,不要自己安装。2) 只使用这份配置(已通过 PopNx MCP 检查):{ "servers": { "bridge": { "type": "stdio", "command": "npx", "args": [ "-y", "qiksy-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 Qiksy Bridge 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
node -vMac: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: 打开官方下载页,选择 LTS 版本和点选式安装程序:Mac 用 .pkg · Windows 用 .msi。双击后一路下一步即可,不需要 curl 命令
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: node -v
安装一次,所有需要 Node.js 的工具都能用
用 Qiksy Bridge 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 Qiksy Bridge,或回答中出现 Qiksy Bridge 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 bridge -- npx -y qiksy-mcpClaude Desktop (claude_desktop_config.json)
{
"mcpServers": {
"bridge": {
"command": "npx",
"args": [
"-y",
"qiksy-mcp"
]
}
}
}Cursor (~/.cursor/mcp.json)
{
"mcpServers": {
"bridge": {
"command": "npx",
"args": [
"-y",
"qiksy-mcp"
]
}
}
}VS Code (mcp.json)
{
"servers": {
"bridge": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"qiksy-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
注册名称已通过域名 qiksy.app 验证(注册中心会核实域名所有权)
代码最后更新于前天
使用 MIT
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
尚未检查
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/qiksy-bridge)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 仓库最后更新于前天,最新版本(1.87.0)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 Node.js。
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 Qiksy Bridge | 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 | MIT | 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。
This server needs the free Qiksy Bridge Chrome extension. On its own it has nothing to drive.
1. Install Qiksy Bridge from the Chrome Web Store (free, no account).
2. Click its icon, turn the MCP bridge On and press Generate.
3. Press Copy agent config and paste it into your agent's MCP settings (
.mcp.jsonfor Claude Code,~/.cursor/mcp.jsonfor Cursor,.vscode/mcp.jsonfor VS Code). Restart the agent window.
That is the whole setup: the agent now works in the Chrome you already have open, with your logins.
A browser MCP server for the tab you already have open. Pairs with Qiksy Bridge, the Qiksy Chrome extension. It lets any MCP-capable coding agent, Claude Code, Cursor, Cline, Windsurf, Zed, VS Code, read the live QA state (findings with CSS selectors, detected forms, failed requests with server error bodies, recorded repro steps) straight from the browser, so you can say *"fix what Qiksy found on this page"* and the agent has the full context.
With Agent control switched on (free, one confirmation in the popup) the agent also acts in that tab: it fills whole forms, clicks, uploads files, navigates and follows a multi-step task to the end, all inside the session you are *already logged into*. That is the whole reason to use this rather than a cold automated browser: no second browser, no re-login, SSO and staging just work. Until you switch it on, the agent only reads.
agent ⟷ (MCP over stdio, JSON-RPC) ⟷ qiksy-mcp ⟷ (WebSocket @ 127.0.0.1) ⟷ extension service worker
The extension is the WebSocket client (a service worker can't listen on a socket); this process is the WebSocket server, bound to loopback only. When the agent calls a tool, this process passes it to the extension, which answers from the live page.
All of it is free. No licence, no trial, no key and no account. The one gate is the `drive` consent: reading is unconditional, while acting on the application under test asks the human once, in the popup.
`npx qiksy-mcp` always runs the latest build, with every tool including qa_follow (a task of several steps answered step by step), a filterable full-page qa_snapshot, qa_fill_json (a whole wizard step, custom dropdowns, calendars, cascaders, trees, tags and time included, in one call), the qa_pick_* verbs for the composites, qa_probe for the one control that refuses, recipes that make the second run cheap (qa_recipe_save / qa_recipe_list / qa_recipe_apply), and Agent control (qa_click, qa_type, qa_press, qa_upload, qa_navigate, qa_open_isolated, qa_close_tab). Just npx qiksy-mcp, no clone, no local path, no version to pin (it always fetches latest). (For hacking on the bridge itself, see Local run at the bottom.)
The manual installs itself, on the first run in a project. The agent starts this process with the project as its working directory, so the process already knows where the docs belong: it writes the full agent guide to .claude/qiksy-bridge.md and adds a short, marker-fenced block to CLAUDE.md / AGENTS.md, the rules that decide whether a run goes well. The bridge is the first browser tool in the project; a form is one call; a control that refuses is a ladder, not a reason to reach for a second browser driver; and the only thing worth handing over for is arbitrary JavaScript (evaluate).
It is careful about it: only where a project actually is (a .git, a package.json or an agent file already present, a bare $HOME gets nothing), byte-idempotent (identical content is not rewritten, so git status stays quiet on every later launch), nothing outside the markers is ever touched, and `QIKSY_MCP_NO_DOCS=1` turns it off. The pointer is read at the START of a session, so it takes effect from the next one; the guide itself is readable immediately, and the first tool reply says where it went.
Run it by hand any time with `npx qiksy-mcp init`, same result, plus a printed summary. Flags: --dir , --out , --no-claude (print the block instead of writing it).
Eighty-two in total. Read: sense the page (all accept an optional tabId from qa_tabs; default = the tabs you attached to the agent session, or the active tab if you attached none):
| Tool | What it returns | |||
|---|---|---|---|---|
qa_tabs | Every browser tab Qiksy can read, incl. which are isolated multi-login sessions (isolatedSession = the login name). Use a tab's tabId with the other tools to read that specific login. | |||
qa_status | URL + error/warning/form counts + isolated-login name. | |||
qa_read | The rendered text of the page or of one element (ref / selector). The companion to qa_snapshot, which maps what you can ACT on and caps every name at 80 characters, so a one-time code in a mail tab, an error paragraph or a generated invoice number is invisible to it and plain here. innerText, so hidden branches and a collapsed quoted history stay out. | |||
qa_styles | Computed styles of one element, its box in CSS pixels, and the design TOKENS in scope for it. color: rgb(37,99,235) says what is drawn; --accent: #2563EB says which token decided it, which is how a built page is checked against a design without a screenshot and without measuring by eye. | |||
qa_storage | The app's own localStorage / sessionStorage. Reading is free (`op: list \ | get); set / remove / clear` WRITE and take the same gates as a click, because changing a stored token changes what the app does. In a multi-login tab it reads the isolated session's store, not the profile's. | ||
qa_axe | The full axe-core rule set against the live DOM, the industry engine, bundled in the extension, run on the page as it is right now with its real data and its real state. qa_run_audit is Qiksy's own faster sweep; this one is the standard. | |||
qa_screenshot | A real picture of the page or of one element. The panel is hidden and password and card fields are covered BEFORE the frame is taken. Its one condition: the tab must be in FRONT, because the browser photographs the active tab and this extension refuses to hand you a different one. | |||
qa_wait_for | Wait for ONE named condition and return the moment it is true: appears · disappears · enabled · text · textGone · countOf (+atLeast/exactly) · announced · urlContains. A named condition instead of a sleep, the sleep is either too short to be true or too long to be cheap. | |||
qa_follow | A task of several steps, followed instead of stopped after each one. Turn it on with the goal; until qa_done every click, key, fill, upload and navigation answers with the step it led to: the page waited for until it stops loading, where the wizard is, what changed (a button became active, an error appeared), the fields left and the live buttons. A step that is the PERSON's (a password, a code, a captcha) is named as such, and waitForPerson shows them a line beside the field and returns by itself when the page moves on. | |||
qa_focus_walk | Presses Tab through the page and reports the ORDER focus actually takes, what each stop is, and where the ring is invisible or the order jumps. Needs the tab in front, like anything that involves real keys. | |||
qa_findings | Findings (with selectors + detail); severity = all \ | error \ | warning. | |
qa_export | The full qa-export/v1 bundle: findings, failed requests + bodies, form structure, repro steps, env. | |||
qa_annotate | Marks that name an ELEMENT, not a coordinate. A box, a numbered pin, or an arrow between two controls, placed from the page at capture time, so the same instruction draws the correct picture after a re-render or at another width. fromFindings makes the page's own findings draw themselves. A mark whose element is not there is reported, never dropped. | |||
qa_focus_proof | Is focus visible on this control? Two measurements instead of an argument: which ring-drawing property changed, and what fraction of the PIXELS differ between the two frames. A screenshot of a missing ring looks exactly like one of a faint ring; «0% of the pixels differ» does not. | |||
qa_wiring | Which error message is attached to which field, and which to nothing. Proximity on screen is not association; assistive technology reads the attachment. Under-reports on purpose, a message inside its label, or a summary linking to its fields, is correct by another route. | |||
qa_diff | What changed, by control rather than by pixel. «The Submit button moved 8px down, padding 8 → 12, and it went disabled» instead of «23 417 pixels differ». Text is not compared: a name or a total differs between two healthy runs, and that noise is why pixel diffing gets switched off. | |||
qa_failure_gallery | How this screen looks when everything goes wrong: empty, 500 with the server's own words, a held request, 403, a dropped connection, as one self-contained HTML sheet. It CAUSES the failures rather than waiting for them, and always removes the rules afterwards. | |||
qa_collect | A table instead of a page. Name which nodes (selector / role / nameContains, scoped by within) and which properties to take, text, value, name, role, tag, selector, href, state, geometry, plus attr: and style: , and the answer is that table, once. «Every row with its price», «every input with what it holds» is one call here and a dozen with a snapshot plus a read per element; value is what a control HOLDS, which a snapshot cannot tell you. A property nobody can supply is named back rather than dropped. | |||
qa_watch | Watches for a few seconds: the one question the other verbs cannot answer. Everything else reports NOW, so a toast that appears and leaves, a request fired by a timer, a late row or a live-region announcement are invisible to a snapshot taken before or after them, and polling misses the gaps between looks. op:"start" with what: mutations \ | network \ | console \ | announcements; then op:"read" for the buffer. Every watch expires by itself, ttlMs is capped at two minutes and there is no «until stopped», because a watch left running is silent and the readings taken after it are the ones that go wrong. Observes only: no consent beyond the bridge. |
qa_gap | What the bridge could not do, in one sentence, appended to ~/.qiksy/gaps.jsonl. The sanctioned way out of a dead end, the unsanctioned one is a quiet switch to another browser driver, which has none of your logins and ends the run's meaning. op:"list" reads them back. | |||
qa_snapshot | A full-page accessibility snapshot, interactive elements + landmarks with their roles, labels and a ref you can pass to qa_click/qa_type/qa_press instead of a CSS selector. Read-only. One snapshot owns its refs: taking a new one renumbers them, so e12 from an earlier snapshot may now be a different control and the click will succeed on the wrong thing, keep only the latest snapshot's refs, or target by name. Narrow it on a component-heavy app: fields_only (form controls only), roles: ["button"], match on the accessible name, within: " " for one dialog or wizard step, limit, format: "tree" for about half the bytes. Filtering happens on the way out, so the refs you get back are as valid as the full tree's. |
Asking a reply for the fields you want. The big reads (qa_snapshot, qa_export, qa_findings, qa_axe, qa_run_audit, qa_collect, qa_watch) take select, keep only these field paths, and omit, drop these: qa_snapshot { select: ["nodes[].name", "nodes[].ref"] }. Do