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

Annotations already establish readOnly/idempotent/non-destructive/closed-world, so the safety profile is covered. Beyond that the description adds real behavior: newest-first ordering, that overdue pages may return fewer rows than limit, the cursor-termination rule, and that 'overdue' is computed from invoice date plus payment terms. Return shape is only loosely implied by the field list.

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 front-loaded, but the view enumeration is stated twice ('Active, paid or all' and again in the next sentence) and partly duplicates the schema's enum. The fragment style ('Active, paid or all.') reads clipped rather than purposeful.

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?

With no output schema, the description compensates by naming the returned fields and the pagination contract, which is what an agent needs to drive the cursor loop. A brief note on maximum page size (limit caps at 50) and on status filtering vs the view parameter would close the remaining gap.

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 80%, so baseline is 3, and the description adds meaning beyond it: view values are defined operationally (active = not paid in full; overdue = active and past due, derived from date + terms), and search is stated to match client name or invoice number. Only limit's bounds and the client_id hint are left entirely to the schema, which already documents them.

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 opening states the resource (invoices), the fields returned (client, total, amount paid, balance due, due date, overdue flag) and ordering ('newest first'), which is far more specific than the title. It does not explicitly contrast with get_invoice or list_estimates, but the list-vs-get distinction is unambiguous given the sibling names.

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?

It explains each view's selection criteria ('active (not paid in full; default)', 'overdue keeps the active ones past their due date', paid, all), which effectively tells the agent when each mode applies. It also warns to follow next_cursor until null. What's missing is routing guidance for search vs client_id and no pointer to get_invoice for single-record detail.

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