Search Term Data Pull
Pure data-extraction layer that feeds every search-term workflow with one consistent pull.
or just say “pull search term data for <brand>”
Comes with the plugin, one install activates every skill. See the install flow. Needs brand setup first.
Negation, harvest, and phrase negative discovery all need the same underlying pull (window performance, three levels of lifetime history, existing negatives, and manual ASIN targets, joined and masked the same way), and building that by hand for every analysis wastes time before the real judgment work even starts.
One structured data artifact, already masked against existing negatives and manual targets and joined at three granularities, that every downstream search term skill reads directly instead of re-deriving its own version. No analysis here, only the clean pull the other skills build on.
What your agent can do
- 01
Pull the window search-term report for the requested date range
- 02
Join lifetime performance at three granularity levels
- 03
Include existing negatives and manual ASIN targets in the same pull
- 04
Output a structured artifact: no judgment or recommendations, data only
Before you run
This skill reads a brand’s context to calibrate its output. Build it once per brand, then every teammate runs on the same shared context.
/mixshift-ai:mx-brand-context <brand>How brand setup works →- Reporting window: the start and end dates the search term report is pulled for
- Spend floor and minimum keep-orders: the thresholds that decide which window rows survive
- Campaign naming pattern: the item-group taxonomy used to map each campaign after the query runs
- Existing negatives and protected terms: the exclusion mask applied before any row reaches a downstream skill
As the account’s item-group taxonomy and naming pattern fill in, fewer campaigns get labeled unknown, so every downstream search term skill works from a more complete corpus while this layer itself stays pure data.