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

Annotations already cover safety (readOnly/idempotent/non-destructive), yet the description adds real behavioral context: the break-even verdict logic, that some spend cannot be split by product, and what labelExport reports. It loses a point because it internally contradicts itself on the default window — 'the last 90 days' vs the schema's 'omit both to read the default 30-day period' — which leaves the agent unsure of the true default.

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?

One dense paragraph: purpose first, then return fields, then the labelExport behavior, then date override and filters. Every sentence carries information, though the long mid-paragraph run-on describing filters could be broken up for faster scanning.

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?

No output schema, so the description correctly carries the return-value burden (verdict buckets, spend, Google-reported conversion value, ROAS, unsplittable spend, labelExport flag), and it documents date override and filters. The main gap is the 90-day vs 30-day default discrepancy and the thin coverage of the remaining filter params, but for an 11-param read-only analytics tool it is close to complete.

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 only 18%, so the description must compensate, and it largely does: it names category, collectionId, label (with the 'exact values from a first call's facets' hint), inStockOnly, newOnly, bucket, search, and sort/dir, and even enumerates bucket values and the sort field contributionMargin. It stops short of syntax/format detail for the remaining params, and dir's semantics are only implied by 'sort/dir reorder it'.

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+resource+scope: Google Shopping/Performance Max spend per Shopify product judged against each product's own break-even ROAS, with the resulting earn/break-even/lose verdicts. It is immediately distinguishable from siblings like get_product_performance or get_contribution_breakdown because it names the channel, the adjudication logic, and the outputs (spend, conversion value, ROAS, unsplittable spend).

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?

Explicit when-to-use ('any question about which products to push or cut on Google, Shopping or Performance Max ROAS by product, or product-level ad profitability'), an exclusion ('Search campaigns are not included'), and a worked routing example ('my best products losing money on Google in category X' = bucket [loses], category [X], sort contributionMargin). Alternatives are implicitly excluded by channel scoping.

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