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

Annotations already declare readOnlyHint=true and openWorldHint=false, so the safety profile is covered; the description adds real substance on top by disclosing the comparison windows, the frequency thresholds (3 cold traffic, 6 retargeting), and the decision rule behind each verdict. It does not state the account/scope boundary (whose account, whether all ads or only active ones) or any cost/latency expectations for the scan.

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?

Front-loaded with what is compared, then the thresholds and verdict taxonomy, then the call-to-action. Every clause carries information — no restatement of the title, no filler. The long opening sentence is dense but each sub-clause adds a distinct signal.

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?

With no output schema, the description carries the burden of describing returns, and it does so by naming the metrics and the four verdict values, which is enough to act on. The gap is that only "fatigued" is given a precise rule; "watch" and "healthy" remain undefined, leaving an agent to guess at boundary interpretation.

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?

The tool takes zero parameters, so the baseline applies and there is nothing for the description to disambiguate in the schema. The description correctly avoids inventing parameter behavior and instead spends its words on computation semantics.

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 (compares) and a precisely scoped resource: every delivering ad's last 7 complete days versus the prior 7. It also enumerates the computed signals (frequency, link CTR change, CPA change) and the four possible verdicts, so the agent knows exactly what the tool returns conceptually.

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?

Gives an explicit trigger: "Call it before recommending new creative or moving budget between ads." That is a clear when-to-use condition tied to a downstream decision. It does not name or exclude siblings such as get_ad_insights or get_account_health, so the agent must infer why this differs from a general performance report.

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