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

Annotations already declare readOnly/idempotent/non-destructive, so the safety profile is covered. The description adds genuine domain behavior beyond that: withholding is not the final tax liability, the 37% threshold above $1 million in annual supplemental pay, and that a separately-paid bonus additionally returns the result under the other federal method for comparison. It does not discuss edge cases like zero-salary or state-specific exceptions, keeping it at 4 rather than 5.

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 the core computation and expected result, then caveats, then alternatives — a sensible ordering with no filler sentences. It is dense but each clause carries information; only the closing comparison note feels slightly tacked on.

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 9-parameter calculator with full schema coverage, an output schema, and complete annotations, the description supplies the remaining context an agent needs: scope (50 states + DC), the domain caveat that withholding ≠ final tax, and sibling disambiguation. Nothing required to select or invoke it correctly is missing.

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 100% schema description coverage the baseline is 3, but the description adds substantive meaning that the schema only implies: it explains the two federal methods and the interaction with the $1M/37% cap, which maps directly to federal_method and earlier_supplemental_pay. It does not elaborate on filing_status, pay_frequency, or w4_step2_checkbox, so it stops short of 5.

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?

Starts with a specific verb and resource ('Calculate how much tax is withheld from a US bonus in 2026 and what is left') and enumerates exactly which components are computed: federal (flat 22% or combined, 37% above $1M), Social Security, Medicare, and state bonus rules. This is unmistakably distinct from siblings like us_paycheck_take_home or overtime_pay.

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?

Gives a literal user-phrasing trigger ('how much of my $X bonus will I keep in <state>?') plus explicit routing: use us_paycheck_take_home for regular pay, salary_increase for a raise. When-to-use, when-not-to-use and alternatives are all present.

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