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

Annotations already declare this a read-only, idempotent, non-destructive, closed-world calculation, so the safety profile is covered. The description adds genuinely new behavioral context: results are a pre-tax estimate, the employer top-up portion is not guaranteed tax-free, and non-Guangdong output is explicitly an unverified approximation. It does not discuss rounding/precision of the figures, which is the remaining gap.

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?

Three compact sentences: what it computes, the regional rule split with an actionable instruction, then the tax caveat. Everything is front-loaded and nothing is redundant, though the middle sentence is dense enough that an agent may need a second pass to parse the conditional rules.

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 8 parameters, one required, a full output schema, and complete annotations, the description covers the open questions an agent would have: regional coverage limits, which fields to override, and the tax disclaimer. Only minor gaps remain, such as whether salary inputs are monthly gross and how the employer differential is computed.

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 100%, so the baseline is 3. The description nevertheless reinforces the two most error-prone distinctions outside the schema text — that local reward leave days are not the same as fund-approved allowance days, and that non-Guangdong users must fill both manually — which meaningfully reduces misuse.

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 opens with a specific verb-plus-resource statement ('分开测算基金生育津贴、单位补差和基金不支付期间的假期工资') and enumerates the three components that are computed separately. Maternity benefits are unmistakably distinct from the other calculate_* siblings (income tax, pension, severance, take-home pay), so an agent can select it without ambiguity.

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 clearly states the accuracy boundary: Guangdong follows verified rules (80 days reward leave excluded from allowance, +30 days for complicated delivery), while other regions are only a national-base trial run requiring the user to hand-enter local reward days and fund-approved days. It does not name a sibling tool to use instead for other regions, so the routing guidance stops one step short of a 5.

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