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

Annotations already mark read-only and idempotent; the description goes further with no purchase/persistence/notification, rate-limit sharing, partial-answer flags, and link handoff semantics. This is exactly the behavioral context needed beyond annotations.

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?

Dense, front-loaded 'Computation only' and no filler; the prose organizes inputs, outputs, and constraints in a logical order. It is long but every clause carries a distinct behavioral or semantic fact.

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?

Despite no output schema, it enumerates candidate fields, page-link semantics, partial results, and compute limits, which is strong coverage for a complex 11-param tool. It does not map each input mode to output kind, but that is a minor gap.

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?

With 0% schema coverage, the description must compensate, and it gives high-level meaning to constraints, free text, and ownership claims. However it does not explain mode, detail, locale/region, max_candidates, or user_supplied_prices, leaving an agent to infer those from names/enums.

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?

Clearly states the tool does computation-only candidate resolution from typed constraints and lists output components, so it is not a purchase or persistence action. It differentiates from sibling search/compare/list tools by emphasizing no purchase/persistence and candidate semantics. The verb 'resolve' plus the resource is specific.

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?

Provides a strong context cue ('Computation only', bound to one published snapshot, budget constraints) that tells an agent when this is the resolution engine rather than an external action. But it never names siblings or states explicit when-to-use/when-not-to-use conditions, so the agent must infer alternatives from sibling names.

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