`peaka_post__data_projects_projectId_cache` – 1 MCP server offers this tool

What it usually takes

No server marks it as only reading: treat it as able to change data.

From each publisher's own tool list. We never call a tool.

Server	Inputs	Read-only?
Peaka `co.huggingface.ramamatar/peaka` Creates a new cache for a specific table in a catalog. Caching copies data from an external data source into the internal Peaka BigTable store for faster query performance. Not every table can be cached: check `isCacheable` and `supportedCacheTypes` in the List Tables response before creating a cache. `supportedCacheTypes` tells which of the modes below are available for the table (`FULL_REFRESH`, `INCREMENTAL`). The cache can be configured in three modes: - One-time cache: No schedules provided. Data is cached once during creation and never automatically refreshed. - Scheduled incremental updates: Provide `incrementalCacheSchedule` to periodically sync only the changes (inserts, updates, deletes) since the last sync. - Scheduled full refresh + incremental: Provide both `fullRefreshCacheSchedule` and `incrementalCacheSchedule`. Full refresh drops and rebuilds the entire cached table periodically, while incremental updates run more frequently between full refreshes. After	`projectId`,`request_body`	Not stated