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

Annotations already declare readOnly/idempotent/non-destructive, but the description adds substantial behavior beyond them: ordering (oldest first), pagination model with items_total, the fact that date_from/date_to are required and inclusive, the draft-vs-complete reporting meaning, and the amount_home unit being integer cents requiring division by 100. The unit warning in particular prevents a real class of agent error.

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 ordering and scope, then required params, then row shape, then filters, then the anti-pattern warning — a logical progression with no filler sentences. It loses a point for mild duplication of the schema (restating the YYYY-MM-DD inclusive format and repeating the filter list already enumerated in the schema), which lengthens an already dense block.

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 an eight-parameter, paginated read tool with no output schema, the description supplies the return-row anatomy (date, number, status, description, account labels, amount_home in cents, cost centers, source) so the agent knows what comes back without an output schema. It also covers pagination and the correct alternative for anything it deliberately does not do.

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 coverage is 100%, so the baseline is 3; the description earns above that by explaining what the schema only lists — status semantics (draft = provisional, not in reports; complete = booked), search matching description or number, and account_id meaning any booking with a leg on that account. It does not clarify page/per_page bounds beyond what the schema already states, which keeps it short of a 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?

States a specific verb and resource (the journal of company bookings) and pins down its distinguishing shape: chronological, oldest first, one page at a time. It explicitly separates itself from two siblings — export_report 'journal' for the leg-by-leg document and the income statement / balance sheet for totals — so an agent can route without opening a schema.

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 concrete trigger phrases ('what was booked in March', 'walk me through last week's entries', checking a sequence of booking numbers) plus explicit exclusions: never sum across rows yourself, use the income statement or balance sheet for totals, and call export_report with report 'journal' for line-level detail. It also advises asking for the next page only when the user needs it.

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