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

With no annotations provided, the description carries the full burden and discharges it exceptionally. It discloses the six-month vesting period, the fact that nothing accrues day by day before that, the ten-day carry-over cap, the twenty-day total cap, and the meaning of capped_by_carry_over_limit / capped_by_total_limit in the output. It also explains that daily_pay valuation is indicative because relevant daily pay is not necessarily salary/260, and that the calculation is the statutory minimum under the Holidays Act 2003 which more generous agreements can exceed.

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?

The structure is logical and front-loaded: purpose, use cases, critical rules, output hints, then args. The all-caps rule announcement is memorable and the procedural explanation of the caps earns its space. It is long, and phrasings like 'This surprises people who assume leave builds up from day one' and 'because "why is my balance not what I counted" is almost always one of those two' are flavorful but not strictly necessary — minor trimming would tighten it without losing information.

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?

Given high complexity — six parameters, 0% schema coverage, no annotations, and no output schema — the description is remarkably complete. It covers the statutory rules, the two caps and the flags that report them, the valuation caveat, the contractual-days escape hatch, and a Returns summary listing all output components. The 'Returns:' paragraph is brief, but with no output schema the description still names the key result fields and their semantics.

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 document all six parameters itself, and it does. Each arg gets a plain-language definition that adds meaning beyond type and title: start_date gets an ISO format requirement, as_at gets its default behavior ('Defaults to today'), carry_over_days is clarified as 'before the cap', and extra_contractual_days is tied back to the statutory-minimum explanation in the prose. daily_pay's optionality and purpose are both 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 opening sentence, 'Work out a New Zealand sick leave entitlement and the balance left,' is a specific verb+resource statement that immediately establishes scope. The 'Use for:' query list ('how much sick leave do I get NZ', 'sick leave entitlement', 'can I carry over sick leave') further pins down intent and distinguishes it from calculators like nz_annual_leave_calculator or nz_parental_leave_calculator.

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?

The 'Use for:' block provides six explicit query patterns that should route the agent to this tool, which is clear context. However, the description never names alternatives or gives exclusions — e.g., it doesn't say to use nz_annual_leave_calculator for annual leave or to use the ACC calculator for work-injury income. Clear context, but no explicit when-not or alternative routing.

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