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

Annotations already declare the safety profile (destructiveHint=true, idempotentHint=true, openWorldHint=false), and the description is consistent with them via 'create or replace'. It adds genuine domain rules — AI usage must be measured or estimated, amounts and rates are never invented — but never states what the replace path overwrites or whether existing linked records are affected, which matters for a destructive tool.

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?

Five short sentences that are front-loaded with the primary action and each carry a distinct rule (item_id, expense_record routing, kinds, minor units, measurement honesty). Some fragments read telegraphically and the kind list restates the enum, but there is little actual waste.

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 destructive, idempotent mutation with 13 parameters, 7 of them required, 15% schema coverage and no output schema, the description leaves major gaps: no return-value expectations, no overwrite semantics, no note on which fields are required or how request_id drives the idempotent retry, and no mention of the business_id scoping that the schema touches only briefly.

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 only 15%, so the description must carry parameter meaning and it largely does not: quantity, unit, client_id, reference, cost_minor, rate_minor, business_id and request_id get no mention. It does clarify `item_id` usage, that the price is in minor units (`amount_minor`), and that `measurement` applies only to AI usage — useful but far short of compensating for 13 mostly undocumented parameters.

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 with a specific verb+resource ('Create or replace unreserved, uninvoiced work') and enumerates the supported kinds (time, reimbursable expense, AI usage), which maps directly to the `kind` enum. It also distinguishes itself from the sibling `expense_record` for linked expenses. The 'replace' half of the verb is not defined (replace what key? by item_id or request_id?), which keeps it short of a 5.

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?

Names a concrete boundary ('unreserved, uninvoiced work') and routes a specific case to an alternative tool ('Linked expenses must be edited with expense_record'). It does not cover other plausible siblings such as timer_start/timer_stop for capturing time, so the alternative-routing is partial rather than complete.

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