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

Beyond annotations it discloses that each call consumes one token, that a task is created and results stored server-side, and that the call is not purely read-only — consistent with readOnlyHint=false and destructiveHint=false. The statement that it does not call external ad accounts or require login sits in mild tension with openWorldHint=true, but the added operational context (idempotency recovery, skip-with-reason behavior) is genuinely useful.

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?

The purpose and the non-mutation caveat are front-loaded, which is good. However, the English idempotency paragraph runs long and much of it restates what the schema's request_idempotency_key description already says, and the mixed-language boilerplate about tasks and tokens dilutes the readable core.

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 read-and-recommend tool with no output schema, the description should sketch the shape of the returned suggestions (candidate type, reason codes, ordering). It explains only that skipped items come back with a reason, leaving the positive-case return values undescribed. The non-mutation guarantee and idempotency behavior are otherwise adequately covered.

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 coverage is 100%, so the schema already defines items, policy thresholds, and the optional idempotency key. The description adds behavioral meaning — empty items yields empty suggestions, unmet thresholds suppress a candidate rather than erroring — but does not clarify how each policy field maps to a specific skip reason or candidate type. A 3 is the correct baseline here.

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?

The opening sentence names a concrete verb and input set (keyword, search term, creative, bid performance) plus the policy criteria, and states the output is improvement candidates. It is clearly distinct from the reporting/reconciliation siblings, though it does not name any alternative tool.

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?

It gives useful conditions — items missing clicks, attribution window, or learning-period thresholds are skipped with a reason, and the result is advisory only, with real changes done separately by the customer. There is no explicit guidance on when to prefer this over a sibling tool or what preconditions the caller must satisfy before supplying data.

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