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

Annotations already declare readOnly, idempotent, non-destructive and closed-world, so the safety profile is covered. The description adds genuinely useful data semantics beyond that: the $100,000 threshold that explains coverage gaps, and the warning that 'Missing lists are not zero; this is not total payroll' — a correctness caveat an agent would otherwise get wrong. Return shape and pagination are not described.

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?

Four compact clauses, front-loaded with the scope and threshold, then the caveat, then the two usage instructions. No sentence is filler, though the telegraphic first clause is slightly harder to parse than a plain verb-first sentence would be.

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 two-optional-parameter tool with no output schema, the description covers purpose, scope caveats, a mode switch and a sibling route. It still omits the shape/size of the returned pay table and any pagination or result-limit behavior, which the agent cannot learn from a missing output schema or from the bare parameter definitions.

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 0%, so the description must carry both parameters, and it largely does: 'by employer and calendar year' maps employer and year to filter semantics, and 'Omit employer to list employers' explains the optional employer parameter's special omitted-mode behavior. Year's expected format (YYYY) is left to the schema pattern rather than stated.

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 description names the resource precisely — published pay above $100,000 broken down by employer and calendar year, with people, common titles, overtime and severance — so an agent knows exactly what data set this returns. The verb is implicit (carried by the name get_pay) rather than stated, and the first clause is a dense noun list, but the scope is unambiguous and it explicitly separates itself from the individual-level sibling ('Individuals use search_records').

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 a concrete mode switch ('Omit employer to list employers') and names an alternative tool with the condition that selects it ('Individuals use search_records'). That is real when-to-use routing, though the boundary against other aggregate siblings like get_totals or get_budget is not addressed.

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