google_ads_generate_keyword_historical_metrics

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations provided, the description carries the full burden, and it delivers: 'read-only planless RPC' explicitly discloses that the call mutates nothing and requires no Keyword Planner plan, and 'computed 3-month and YoY changes' discloses that the server pre-processes data rather than returning only raw series. It does not cover rate limits, permission requirements, or failure behavior, but for a read-only tool the central state-modification question is answered. The read-only disclosure adds context the schema alone could not convey.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

Three sentences with zero filler: action in sentence one, compact return-value enumeration in sentence two, and the read-only/planless qualification in sentence three. The most selection-relevant information (what it fetches and returns) is front-loaded ahead of the safety classification.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 12-parameter tool with no output schema, the explicit return-value enumeration is the critical missing piece, and the description provides it, naming every output group an agent would want to map to downstream use. The behavior classification (read-only, planless) and the history-length caveat ('up to 48 months', with 24 enabling YoY as stated in the schema) round out the operational picture. It stops short of routing among the three keyword siblings and omits response-size guidance, which the schema's 50,000-point guard only partially covers.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 83%, with only network (an enum with self-explanatory values) and includeAdultKeywords (a self-describing boolean) lacking descriptions, so the schema already does the heavy lifting. The description's mention of 'close variants' and 'up to 48 months' reinforces the keywords and historyMonths parameters but adds no new syntax or format details. Baseline 3 is correct because the description adds marginal value over an already well-documented schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('Get Keyword Planner search-volume history') and enumerates the exact return payload (average monthly searches, monthly volumes up to 48 months, latest volume, 3-month/YoY changes, competition, CPC/bid ranges, close variants, device totals). This clearly distinguishes it from sibling keyword tools: generate_keyword_ideas produces candidate keywords and generate_keyword_forecast_metrics produces future projections, while this one retrieves historical volume. The 'read-only planless RPC' clause further separates it from the GAQL and run_readonly_rpc siblings.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description makes the tool's scope clear (historical search-volume data, not ideas or forecasts) so an agent can infer when to choose it, but it never names an alternative or states an exclusion. With close siblings like google_ads_generate_keyword_ideas and google_ads_generate_keyword_forecast_metrics, explicit 'use this for historical data, that for future estimates' routing would materially improve selection accuracy. As written, usage guidance is implied rather than stated.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.