A modern pytrends alternative for AI agents

pytrends scrapes Google Trends and breaks when Google changes its frontend. Trends MCP is the AI-native replacement: structured trend data via the Model Context Protocol, with absolute volume estimates, multi-platform queries, and no maintenance overhead.

Tools for this workflow

`get_trends`

Get the same historical time series pytrends returns - but with absolute volume estimates added, no scraping fragility, and no Python script required.

get_trends(keyword='electric vehicles', source='google search', data_mode='weekly')

`get_growth`

Replace pytrends' interest_over_time with structured growth metrics across 15+ platforms in one call - pytrends is Google-only and returns relative data only.

get_growth(keyword='electric vehicles', source='google search, youtube, reddit', percent_growth=['1Y', '3M'])

`get_ranked_trends`

Discover top-trending keywords with growth rankings - a capability pytrends offers only partially via related_queries, with no cross-platform support.

get_ranked_trends(source='google search', sort='yoy_pct_change', limit=30)

`get_top_trends`

Pull live breakout trending searches - the equivalent of pytrends' trending_searches but delivered as structured JSON your AI can reason over directly.

get_top_trends(type='Google Trends', limit=25)

FAQ

Common questions

What is pytrends?

pytrends is an unofficial Python library that reverse-engineers the Google Trends website to fetch relative search interest data. It is widely used but fragile - it breaks when Google updates its site, has strict rate limits, and only returns relative (0-100) interest scores with no absolute volume.

Why switch from pytrends to Trends MCP?

Three main reasons: (1) Trends MCP returns absolute volume estimates alongside relative interest - pytrends only gives relative data. (2) Trends MCP queries 15+ platforms in one call - pytrends is Google-only. (3) Trends MCP is designed for AI agents - it returns structured JSON your AI can reason over without any Python scripting.

Does Trends MCP work if I am not using Python?

Yes. Trends MCP is client-agnostic. Any MCP-compatible AI (Claude, Cursor, VS Code, Windsurf, ChatGPT) can call it without writing any code. The AI handles the tool calls directly.

Can I get the same data pytrends returns via Trends MCP?

Yes - interest_over_time, related_topics, trending_searches, and regional breakdown are all available. Plus additional signals pytrends cannot access: TikTok, Reddit, YouTube, Amazon, Wikipedia, news sentiment, web traffic, and app downloads.

get_trends

get_growth

get_ranked_trends

get_top_trends

Common questions

Start building with live trend data

`get_trends`

`get_growth`

`get_ranked_trends`

`get_top_trends`