No user-submitted related servers found.
Scored across 41 tools
Most tools target a distinct resource or action, and the descriptions clearly separate pairs like pause_keywords vs pause_ad_groups and add_negative_keywords vs add_shared_negative_keywords. However, the four performance-variant tools, the preview_campaign/forecast_campaign pair, and the governance tools (policy_status, list_capabilities, explain_capability) require careful reading to pick correctly, and the generic mutate/run_gaql escape hatches broadly overlap with the specific tools.
Almost all tool names are snake_case with an imperative verb first, following a list_/create_/preview_/pause_/remove_/promote_ pattern. The main deviations are the noun-style report tools like keyword_performance, ad_asset_performance, geo_performance, search_terms, and conversion_actions, which are consistent in style but not verb_noun. Overall the naming is predictable and readable.
At 41 tools, this is well over the 25-tool threshold and makes for a heavy selection surface that will consume significant agent context. The broad Google Ads scope justifies much of the breadth, but clusters like the four performance reports, the capability/proposal machinery, and the generic mutate path could be consolidated. It is capable but oversized.
The toolset covers the campaign lifecycle, keyword/ad group/ad operations, negative keywords, drafts, audits, forecasting, performance analysis, and governance, with mutate and run_gaql as escape hatches so there are no dead ends. A few common operations such as pausing/toggling campaigns or creating ad groups are not first-class tools and must go through mutate, and shared-list creation is left to raw API calls. These are workable gaps rather than blocking ones.
