yandex-wordstat_skill

This skill analyzes search demand and keyword statistics via Yandex Wordstat, clarifies region, verifies intent, and exports actionable insights.

Python

30

GitHub Stars

2

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill artwist-polyakov/polyakov-claude-skills --skill yandex-wordstat

.gitignore94 B
SKILL.md10.7 KB

Overview

This skill analyzes search demand using the Yandex Wordstat API to build keyword lists, measure frequency, and detect seasonality and regional demand. It delivers up to 2000 top queries, associations, trend dynamics, and CSV export for downstream processing. The workflow emphasizes intent verification and region-aware filtering to separate target and non-target queries.

How this skill works

The skill queries Yandex Wordstat to retrieve two result types: top requests (phrases containing your words) and associations (semantically related queries). It supports regional filters, device filters, and time-series dynamics. For every promising query the workflow requires a WebSearch check to verify commercial intent before marking it as target.

When to use it

When researching keyword demand and building a semantic core for paid search or SEO
When you need regional distribution or city-level demand for a product or service
When you want trend data (daily/weekly/monthly) or seasonality analysis
When exporting large keyword sets (up to 2000) to CSV for further processing
When expanding keywords with associations but needing manual intent verification

Best practices

Always ask and fix the target region before starting any analysis
Clarify exactly what you sell to filter out accessory or DIY queries
Run WebSearch for each promising query to confirm transactional intent
Use Wordstat operators (quotes, !, -, grouping) to control match type and clean noise
Export large results as CSV and process in chunks to avoid stdout flooding

Example use cases

Build a purchase-focused keyword list for chimney sales in Moscow and exclude accessory searches
Export 2000 queries for an advertiser and segment by top/assoc for manual intent review
Analyze monthly dynamics for a legal service phrase to plan seasonal ad budgets
Get city-level demand breakdown for a product launch using regions_stats.sh
Filter out informational and DIY queries by combining operators and WebSearch verification

FAQ

Set YANDEX_WORDSTAT_TOKEN in the environment config before running any script.

How do I ensure results target buyers and not DIYers?

Clarify the product you sell, use minus operators to exclude DIY terms, and always verify intent via WebSearch for each promising query.

Can I get more than 50 results?

Yes — use the --limit parameter up to 2000 and export the full set to CSV for complete output.