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

Annotations indicate readOnlyHint=false and destructiveHint=false, so this is a mutation (generation) tool. The description adds important behavioral context: it is a job-based operation ('Platform key Job') and the result is a signed download_url via jobs.result_storage_uri, not an immediate file. It also clarifies that tax/deduction amounts are already computed, so the tool does not calculate them. This goes beyond the annotations and helps set expectations for an asynchronous, file-generating process. No contradiction with annotations.

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 description is two concise sentences with no redundancy. It front-loads the core purpose and input, then provides the key result format and a distinguishing note. Every word adds value, and it is appropriately brief for a tool of this complexity. The structure is efficient and easy to scan.

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 the tool has 6 parameters, a nested employees array, and no output schema, the description covers the essential outcome (signed download_url via jobs.result_storage_uri) and the fact that amounts are pre-computed. It also mentions the platform key and absence of account_link_id, which are operational prerequisites. It does not mention potential error conditions or maximum array sizes, but for a generation tool, the essentials are present. The description is largely complete for an agent to call it correctly, though it could benefit from noting any asynchronous polling requirements explicitly.

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 50%, meaning half of the parameters lack descriptions. The tool description only mentions attribution_ym and employees[], ignoring the optional parameters like workplace_name, workplace_address, and business_registration_no. It does not add meaning beyond the schema for the required parameters, nor does it compensate for the undocumented ones. For example, it doesn't explain the relationship between monthly_pay_won and monthly_salary_won beyond what the schema already provides. The description is not sufficient to guide parameter usage given the low schema coverage.

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 clearly states the tool builds a standard payroll register workbook (xlsx) from attribution_ym and employees[]. It names the specific verb (Build), the resource (payroll register workbook), and the inputs. It also distinguishes itself by noting 'not composite merge' and the platform key Job, which differentiates it from siblings like payroll_artifact__payslip_xlsx, though not explicitly. The purpose is unambiguous and specific.

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 description provides no guidance on when to use this tool versus alternatives. It does not mention the sibling payslip_xlsx or any other payroll-related tools. The phrase 'not composite merge' is a hint about method but not about selection. It fails to state conditions for using this tool instead of others, leaving the agent to infer from the name alone. There is no when-to-use or when-not-to-use guidance.

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