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

Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false and openWorldHint=false, so the safety profile needs no restating; the description instead adds genuinely new behavior: the async generation prerequisite, the polling gate, and the fact that errorCount/warningCount are combined totals across two distinct error sources. That said, the annotation bar is already met, so the added behavioral context is useful but sits at a solid rather than exceptional level.

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?

Purpose and use case are front-loaded and every section earns its place, but it is delivered as a single dense paragraph mixing purpose, domain background, return shape, error taxonomy, and async sequencing, which makes it harder to scan than a bulleted structure would. The unexplained abbreviation 'CP' is a minor comprehension tax.

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 fully specifies the return object and each error's fields, plus the semantic distinction between dataCheckErrors (source='export', pre-submission) and channelErrors (source='adapter'/'api_error', post-submission) and the meaning of level, count, statusCounts and listingStatusCounts. Nothing needed to call this correctly or interpret the result is missing.

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 0%, so the description carries the full burden and does: ad_id is 'the id from list_ads', and project_id is described as OPTIONAL with its inference rule ('inferred for a single-project customer') plus the failure mode (project_id_required) and the recovery call (list_projects). Both of the two parameters are fully disambiguated.

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 the validation report for one ad') and immediately scopes it as 'aggregated errors and warnings that block or degrade its feed, with fix suggestions', which separates it from get_ad (config), ad_status (progress), and get_ad_item_report (per-item). It also defines what an 'ad' is, removing ambiguity with marketplace 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?

Explicitly names the use case ('explain why ad products fail validation') and gives the operational precondition chain: data-check errors exist only after a generation has run, so finish the wizard or call repair_ad, and because export is async, poll ad_status until productsRefreshing and productsSubmitting are false before reading. It routes the agent to ad_status, repair_ad, configure_ad_step, list_ads and list_projects by name.

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