yandex-wordstat-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| YANDEX_FOLDER_ID | Yes | Yandex Cloud folder ID. | |
| YANDEX_IAM_TOKEN | No | IAM token to use instead of an API key. Valid for 12 hours; suitable for one-off runs. | |
| YANDEX_SEARCH_API_KEY | No | API key for the Yandex Cloud service account. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
Yandex Wordstat: how often people search for a phrase in Yandex.
top_requests: monthly volume for a phrase plus related queries (keyword research).
frequencies: base and exact volume for up to 100 phrases in one call (compare wordings).
dynamics: how volume changes over time (seasonality, trends).
regions_distribution: where people search for a phrase, with an affinity index.
find_region: look up region IDs by name for the
regionsfilters. Phrases support Wordstat operators: "quotes" fix the word set, !word fixes the form, +word keeps stop words, -word excludes. Data lags real time by several weeks.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| top_requestsB | Search volume for a phrase over the last 30 days, plus the most popular queries containing it and similar ("associated") queries. Use for keyword research. |
| frequenciesA | Base and exact search volume over the last 30 days for a list of phrases in one call.
Use it to check real demand for specific wordings instead of calling top_requests per phrase.
|
| dynamicsA | How search volume for a phrase changes over time. |
| regions_distributionB | Where people search for a phrase over the last 30 days. |
| find_regionA | Find Wordstat region IDs by name. Returns id, name and the path in the region tree. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 5 tools
The main overlap risk is between top_requests and frequencies, since both return 30-day search volume, but the descriptions clearly split them: top_requests for discovery (associated queries) and frequencies for batch base/exact checks. The remaining three tools (dynamics, regions_distribution, find_region) target time series, geography, and region lookup respectively, so intent is easy to infer.
All names use lowercase snake_case, which keeps formatting consistent. Semantically, four are noun-phrases naming the data returned (top_requests, frequencies, dynamics, regions_distribution) while find_region uses a verb_noun form, a minor stylistic deviation that is still readable.
Five tools is a well-scoped surface for a keyword-research API, with no redundant or filler endpoints. Each tool covers a distinct query dimension (volume, time, geography, region resolution).
The surface covers volume, base/exact frequency, trends over time, regional breakdown, and region ID resolution, which is close to full lifecycle for Wordstat research. Minor gaps remain, such as no explicit tool for comparing multiple phrases over time or fetching the full associated-query list separately from volume, but the core workflows are achievable with the given tools.