Google Trends Mcp

io.github.purahmanian由 GitHub 账号 purahmanian 发布v0.1.1更新于 Oct 24, 2026

Google Trends MCP server: interest over time, comparisons, related queries, trending searches

概览

AI 生成的概览

让助手获取 Google Trends 数据:随时间变化的关注度、关键词对比、相关查询、热门搜索和地区关注度。

可以做什么
这个本地 MCP 服务器提供五个基于 Google Trends 的工具:interest_over_time 返回最多五个关键词的每周 0-100 关注度分数;compare_terms 对两到五个关键词做归一化比较并给出胜出者;related_queries 列出相关查询中的热门和上升项;trending_now 返回某国家当天的热门搜索;interest_by_region 按国家或次级地区拆分关注度。它通过 stdio 运行,无需 API 密钥。
适用场景
当你想在 AI 客户端中获得搜索需求信号时使用:查看季节性、比较关键词或品牌关注度、发现上升的相关查询,或了解某国的热门搜索。适合营销、产品调研和内容规划,而非精确的市场测量。
运行要求
需要本地 Node.js 运行环境和 npx,因为它以 npm 包 google-trends-mcp 通过 stdio 安装并运行。未声明需要 API 密钥、账号或环境变量。需要能访问 Google Trends 端点的出站网络。
安装前请注意
它调用 Google Trends 未公开的内部端点,Google 可能随时更改、限流或移除;可能出现 HTTP 429 或被封禁,实际速率约为每分钟几次请求。你查询的搜索词和地区代码会发送给 Google。该项目为非官方,与 Google 无关联。

安装

在 SourceWeft 中

  1. 打开 控制台中的 Google Trends Mcp,将其添加到工作区。
  2. 为需要使用其工具的对话启用该服务。

Desktop only,通过 STDIO。 STDIO 服务会启动本地进程,因此需要 SourceWeft 桌面宿主。

其他 MCP 客户端

参照 仓库 中的启动说明。

README

google-trends-mcp

Free Google Trends data inside any MCP-compatible AI client. No API key required.

[CI] [npm]


UNOFFICIAL ENDPOINT DISCLAIMER

This server communicates with the same internal endpoints that the trends.google.com frontend uses. These endpoints are undocumented and unofficial. Google can change, rate-limit, or remove them at any time without notice. This project is not affiliated with, endorsed by, or sponsored by Google LLC. "Google Trends" is a trademark of Google LLC.

If requests start failing, open an issue. The fix is usually a URL or parameter tweak.


What it does

ToolDescriptionExample prompt
interest_over_timeWeekly interest scores (0-100) for up to 5 terms"Show me interest in 'matcha latte' over the past year in the US"
compare_termsNormalized comparison of 2-5 terms with winner callout"Compare 'coffee', 'tea', and 'matcha' in Canada"
related_queriesTop and rising related search queries for a term"What related queries are rising for 'cold plunge'?"
trending_nowToday's daily trending searches for a country"What's trending on Google in the UK right now?"
interest_by_regionRegional breakdown of interest (country or sub-region)"Which US states search for 'pickleball' most?"

Quick start

No API key is needed. Install and run via npx.

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

json
{  "mcpServers": {    "google-trends": {      "command": "npx",      "args": ["-y", "google-trends-mcp"]    }  }}

Claude Code

bash
claude mcp add google-trends -- npx -y google-trends-mcp

OpenAI Codex CLI

Add to ~/.codex/config.toml:

toml
[mcp_servers.google-trends]command = "npx"args = ["-y", "google-trends-mcp"]

Example conversations

Spotting seasonality:

"Use Google Trends to show me interest in 'pumpkin spice' over the past 5 years worldwide."

The model calls interest_over_time with terms=["pumpkin spice"], timeframe="today 5-y" and narrates the clear September-October spikes each year.

Keyword competition:

"Compare search interest in 'Notion', 'Obsidian', and 'Roam Research' in the US this year."

The model calls compare_terms, reads the averages, and identifies the winner with supporting time-series data.

Opportunity discovery:

"What are the fastest-rising related queries for 'sourdough bread' right now?"

The model calls related_queries and surfaces the rising queries, often revealing adjacent niches and trending subtopics.


Development

bash
# Clone and installgit clone https://github.com/purahmanian/google-trends-mcp.gitcd google-trends-mcpnpm install
# Run tests (no network calls; all HTTP is mocked)npm test
# Build TypeScript to dist/npm run build
# Run the server locally (stdio mode)node dist/index.js

Rate limits and blocking

Google does not publish a rate limit for these endpoints. In practice, a few requests per minute works reliably. If you hit HTTP 429 or get blocked, the server returns a descriptive error message. Waiting 2-5 minutes usually resolves it.


Built by

Built by Puya Ventures LLC. I build custom MCP servers and AI integrations for product teams and researchers. Get in touch: [email protected] | Portfolio: puyarahmanian.com

Part of the Product-Research MCP Suite: keepa-mcp · google-trends-mcp · junglescout-mcp


Privacy

This server runs entirely on your machine. It collects no telemetry and stores no data. The only network calls it makes are to Google Trends endpoints (trends.google.com), sending the search terms and region codes you ask about. No account or API key is involved. See Google's privacy policy: https://policies.google.com/privacy

License

MIT. See LICENSE.

来源:README.md,提交 a505da1

工具

0
SourceWeft 尚未在沙箱中启动过该软件包,因此其工具未知。

版本历史

1
  1. v0.1.1最新Sep 16, 2026

评分

已满足适用检查项的 50%

该评分衡量此收录条目的完整程度,依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价,也不表示服务器安全或当前可用:可连通性是带日期的检查结果,评分也不包含安装量或用户评价。

  • 近 30 天内可连通25 分

    请保持端点能够响应 MCP 握手;SourceWeft 每 30 天探测一次。

  • 已声明认证方式10 分

    请在注册中心条目中声明凭据请求头,或保持端点能够响应探测,以便判定其认证方式。

  • 已列出工具15 分

    请让服务器在无凭据时响应 tools/list,或发布可在沙箱中无凭据启动的软件包。

  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分