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

Annotations only declare readOnlyHint and openWorldHint; the description adds substantial context beyond them: regional API coverage, two distinct hard-refusal preconditions (payroll setup complete, connecting user held payroll admin), the active-only filter, per-country caveats on how employment status is derived, super fund data being Australia-only, and explicit non-disclosure of TFN, bank account and address. This is exactly the extra behavioural context annotations cannot carry.

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?

Front-loaded with what is returned and the region gate, and almost every sentence carries operational consequence. It is dense rather than padded, though the 'still accurate, just not Xero's own word for it' aside and the repeated 'this refuses plainly' framing could be tightened without losing meaning.

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 compensates by enumerating the returned fields, the optional super fund data, the omitted sensitive fields, and the pagination shape. The main gaps an agent would hit — unsupported regions, incomplete payroll setup, absent NZ/UK super data, and partial pages — are all pre-answered.

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 documents all three parameters including includeInactive's semantics. The description adds one genuinely useful fact not in the schema — the 100-records-per-page page size — but says nothing further about page bounds or the organisation parameter.

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?

Opens by naming the exact resource (this organisation's payroll employees) and the specific fields returned (name, employment status, super fund presence, ordinary earnings rate id), which cleanly separates it from siblings like list_xero_payroll_earnings_rates and list_xero_payroll_leave_types. Scope qualifiers (AU/NZ/UK, active-only default) sharpen the identity further.

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 about when it applies (three supported countries), when it refuses and how to report that refusal rather than implying zero employees, how to use includeInactive, and how to handle a full page of 100. It also states outright what the tool cannot answer (payment status, ATO/IRD/HMRC lodgement), which steers the agent away from misusing it. No sibling is named, but no viable alternative exists for this resource.

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