google-search-trends-mcp
Provides Google Search trend data as an MCP tool, with historical series, growth percentages, and live trending searches, no scraping or rate limits.
README
google-search-trends-mcp
The number one Python package for Google Search trend data. Google Search trends as an MCP tool. Plug into Claude, Cursor, or any MCP-compatible AI host. Weekly series, growth percentages, and live Google trending searches.
Powered by trendsmcp.ai, the #1 MCP server for live trend data.
Get your free API key at trendsmcp.ai - 100 free requests per month, no credit card.
?? Full API docs ? trendsmcp.ai/docs
Updated for 2026. Works with Python 3.8 through 3.13.
Use as an MCP tool
Add to your mcp.json (Claude Desktop, Cursor, or any MCP host):
{
"mcpServers": {
"trends": {
"command": "npx",
"args": ["-y", "trendsmcp"],
"env": { "TRENDS_API_KEY": "YOUR_API_KEY" }
}
}
}
Get your free key at trendsmcp.ai.
No scraping. No 429 errors. No proxies.
If you have used pytrends or similar scrapers before, you know the problems: random 429 Too Many Requests blocks, broken pipelines at 2am, time.sleep() hacks, proxy rotation costs, and a library that is now archived because Google explicitly flags scrapers at the protocol level.
trendsmcp is the managed alternative. We run the data infrastructure. You call a REST endpoint.
pytrends alternative for Google Search data
| Scrapers / pytrends | trendsmcp | |
|---|---|---|
| 429 rate limit errors | constant | never |
| Proxy required | often | never |
| Breaks on platform changes | yes, regularly | no |
| Platforms covered | 1 (Google only) | 13 |
| Absolute volume estimates | no | yes |
| Cross-platform growth | no | yes |
| Async support | no | yes |
| Actively maintained | no (archived) | yes |
| Free tier | no | yes, 100 req/month |
Install
pip install google-search-trends-mcp
Zero system dependencies. Python 3.8 or later. Uses httpx under the hood.
Quick start
from google_search_trends_mcp import TrendsMcpClient, SOURCE
client = TrendsMcpClient(api_key="YOUR_API_KEY")
# 5-year weekly time series, no sleep(), no proxies, no 429s
series = client.get_trends(source=SOURCE, keyword="bitcoin")
print(series[0])
# TrendsDataPoint(date='2026-03-28', value=72, keyword='bitcoin', source='google search')
# Period-over-period growth
growth = client.get_growth(
source=SOURCE,
keyword="bitcoin",
percent_growth=["12M", "YTD"],
)
print(growth.results[0])
# GrowthResult(period='3M', growth=14.5, direction='increase', ...)
# What's trending right now
trending = client.get_top_trends(limit=10)
print(trending.data)
# [[1, 'topic one'], [2, 'topic two'], ...]
Async support
import asyncio
from google_search_trends_mcp import AsyncTrendsMcpClient, SOURCE
async def main():
client = AsyncTrendsMcpClient(api_key="YOUR_API_KEY")
series = await client.get_trends(source=SOURCE, keyword="bitcoin")
print(series[0])
asyncio.run(main())
Run multiple platform queries concurrently:
google, youtube, reddit = await asyncio.gather(
client.get_trends(source="google search", keyword="bitcoin"),
client.get_trends(source="youtube", keyword="bitcoin"),
client.get_trends(source="reddit", keyword="bitcoin"),
)
Use cases
- SEO research: track keyword search volume trends across Google Search, Google News, and Google Images before publishing content
- Market research: measure consumer demand signals on Amazon and Google Shopping before entering a product category
- Investment research: monitor Reddit discussion volume, news sentiment, and Wikipedia page view spikes as leading indicators
- Content strategy: find what is growing on YouTube and TikTok before topics peak and competition saturates them
- Competitor tracking: compare brand search volume growth across platforms over custom date ranges
Works with
- Claude (via MCP server at trendsmcp.ai)
- Cursor (via MCP server at trendsmcp.ai)
- ChatGPT (via MCP server at trendsmcp.ai)
- VS Code Copilot (via MCP server at trendsmcp.ai)
- LangChain: pass
TrendsMcpClientoutput directly as tool results or context - LlamaIndex: use trend series as structured data nodes for retrieval
- Pandas: each
get_trends()response converts to a DataFrame in one line
Methods
get_trends(source, keyword, data_mode=None)
Returns a historical time series for a keyword. Defaults to 5 years of weekly data. Pass data_mode="daily" for the last 30 days at daily granularity.
get_growth(source, keyword, percent_growth, data_mode=None)
Calculates percentage growth between two points in time. Pass preset strings or CustomGrowthPeriod objects.
Growth presets: 7D 14D 30D 1M 2M 3M 6M 9M 12M 1Y 18M 24M 2Y 36M 3Y 48M 60M 5Y MTD QTD YTD
get_top_trends(type=None, limit=None)
Returns today's live trending items. Omit type to get all feeds at once.
Available feeds: Google Trends YouTube TikTok Trending Hashtags Reddit Hot Posts Amazon Best Sellers Top Rated App Store Top Free Wikipedia Trending Spotify Top Podcasts X (Twitter) and more.
All 13 supported sources
One API key. One client. All platforms. No separate credentials for each.
| source | What it measures |
|---|---|
"google search" |
Google Search volume |
"google images" |
Google Images search volume |
"google news" |
Google News search volume |
"google shopping" |
Google Shopping purchase intent |
"youtube" |
YouTube search volume |
"tiktok" |
TikTok hashtag volume |
"reddit" |
Reddit mention volume |
"amazon" |
Amazon product search volume |
"wikipedia" |
Wikipedia page views |
"news volume" |
News article mention count |
"news sentiment" |
News sentiment score (positive/negative) |
"npm" |
npm package weekly downloads |
"steam" |
Steam concurrent player count |
All values normalized 0 to 100 on the same scale so you can compare across platforms directly.
Error handling
from google_search_trends_mcp import TrendsMcpClient, TrendsMcpError, SOURCE
client = TrendsMcpClient(api_key="YOUR_API_KEY")
try:
series = client.get_trends(source=SOURCE, keyword="bitcoin")
except TrendsMcpError as e:
print(e.status) # e.g. 429 if you exceed your plan quota
print(e.code) # e.g. "rate_limited"
print(e.message)
Frequently asked questions
Does this scrape Google Search? No. trendsmcp runs managed data infrastructure. Your Python code makes a single authenticated REST call. No scraping, no Selenium, no cookies, no proxies required.
Do I need a Google Search developer account, OAuth token, or platform API key? No. One trendsmcp API key gives you access to all 13 sources.
Will it break when Google Search changes its backend? No. API stability is our responsibility. If something changes upstream, we update the backend. Your code keeps working.
Is there a free tier? Yes, 100 requests per month, no credit card required. Get your key at trendsmcp.ai.
Can I use this in production data pipelines? Yes. The client is stateless, thread-safe, and supports async for concurrent queries across multiple platforms.
Related packages
- trendsmcp - core package, all 13 sources
- youtube-trends-api / youtube-trends-mcp
- reddit-trends-api / reddit-trends-mcp
- google-search-trends-api / google-search-trends-mcp
- amazon-trends-api / amazon-trends-mcp
- tiktok-trends-api / tiktok-trends-mcp
- wikipedia-trends-api / wikipedia-trends-mcp
- npm-trends-api / npm-trends-mcp
- steam-trends-api / steam-trends-mcp
- app-store-trends-api / app-store-trends-mcp
- news-volume-api / news-volume-mcp
- news-sentiment-api / news-sentiment-mcp
Links
Also works as a Python client
Same API key works directly in Python - no MCP host needed.
pip install google-search-trends-mcp
import os
from google_search_trends_mcp import TrendsMcpClient, SOURCE
client = TrendsMcpClient(api_key=os.environ["TRENDSMCP_API_KEY"])
series = client.get_trends(source=SOURCE, keyword="your keyword")
growth = client.get_growth(source=SOURCE, keyword="your keyword", percent_growth=["1M", "3M", "12M"])
top = client.get_top_trends(type="Google Search", limit=10)
Full Python docs: trendsmcp.ai/docs
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。