Google Shopping agent
There is no Shopping board. Pick a shopper phrase. Run source google shopping 3M and 12M, then source google search on the same words. Rising shopping with flat web search is purchase intent. The reverse is research without a cart.
The agent should not invent a Shopping best-seller list. That type does not exist. Start from a phrase a shopper would type.
GSINs are the wrong object. running shoes is the right one.
Answer in those terms: shopping-tab interest versus web search, both 0-100.
The loop
For running shoes, pull 12-month google shopping growth and 12-month google search growth. Say purchase intent versus research.
1. get_growth(keyword="running shoes", source="google shopping", percent_growth=["3M","12M"]) 2. get_growth(keyword="running shoes", source="google search", percent_growth=["3M","12M"]) 3. Answer. Purchase intent vs research.
Run it
Sign upFree. No card. Then ask this loop.
Questions
Board first?
No Shopping board. Start with a phrase.
GTIN?
No. Shopper phrase.
Two sources?
google shopping then google search.
Merchant Center?
Not in the loop.
How do I run this?
Sign up. Free, no card. Then ask Claude, Cursor, or ChatGPT to run the loop on this page.
Other agents
YouTube trending agent TikTok trending hashtags agent Reddit Hot agent