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

Annotations already declare readOnlyHint, idempotentHint, non-destructive, and closed-world, so safety needs no restating. The description adds genuinely non-obvious behavior: which reports the range parameters mean different things for, that general_ledger is paged, that oss needs 'no books needed', that reconciliation returns reconciled/to_match/off, and that cash_forecast omits undated overdue items and names the sibling tools to fix them. It does not mention pagination limits or result sizes explicitly, keeping it short of a 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?

The content is dense and almost every clause carries information, so the length is largely justified by thirteen report types. But it is delivered as four massive semicolon-chained sentences with no list structure, which makes it slow to scan and front-loads only the base-currency note rather than a clear 'runs one of these reports' opener.

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?

There is no output schema, so the description has to carry return-value meaning, and it does so well for many reports (running balances and paging for general_ledger, per-country/per-rate euros for oss, by day and week plus lowest point for cash_forecast, columns for compare). The notable omissions are the shape of limit/offset behavior and any indication of what the simpler reports return, so it falls just short of complete.

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 79% schema coverage the baseline is 3, but the description binds parameters to specific report modes in a way the schema does not — days for cash_forecast, quarter for oss, country for vat_return, budget for budget_variance, compare/periods/by_tracking/tracking for profit_and_loss, and account for general_ledger. The only real gap is limit and offset, which are covered solely by the oblique word 'paged' and carry no schema description.

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 makes clear this is a dispatcher that runs one of thirteen named financial reports, and it explains what each one is (profit_and_loss, balance_sheet, oss, cash_forecast, etc.). This lets an agent map the `report` enum to intent without opening the schema. It does not, however, explicitly distinguish itself from adjacent reporting siblings such as books_export, invoices_report, or purchasing_report, so the sibling differentiation is inferred rather than stated.

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?

For nearly every report it states the applicable parameters and the default behavior (e.g. 'from-to (default: this fiscal year so far)', 'quarter, default the last one ended', 'days ahead (default 30, up to 180)', 'country, default the business's'), which is effectively per-mode when-to-use guidance. It also routes the display case to view_show. It stops short of explicit exclusions ('do not use for X, use sibling Y instead') and never contrasts with the other *_report tools.

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