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

Annotations already declare readOnlyHint=true, destructiveHint=false and openWorldHint=false; the description reinforces this with the concrete list of write operations that are excluded, which is useful context beyond the hint. It also discloses default ordering and the per-row fields. It does not mention pagination behavior or the 100-row ceiling, which is the remaining behavioral gap.

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?

Two dense sentences, front-loaded with the resource and return fields, then the primary filter idiom, then the read-only caveat. Every clause carries information an agent would otherwise have to infer; nothing is restated from the title.

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 9-parameter, zero-required list tool with no output schema, the description covers what rows contain, their order, how to filter, and what the tool cannot do. The gaps — allowed values for the free-text filter params and how limit/paging behaves — are moderate but real given the 44% schema coverage and absence of enums.

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 only 44%, so the description carries real weight and does — it explains the semantics of outstanding (money still owed) and enumerates every filter except limit, telling the agent the filtering surface is combinable. It does not supply valid values for status, docType or paymentState, and with zero enums in the schema those remain guesswork, 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 ("List the company's invoices from the ERP") plus the three returned measures (total, paid, OUTSTANDING) and the ordering (newest first). That is enough for an agent to separate it from erp_order_list, erp_list_contacts, erp_list_attachments and the other list siblings by resource alone.

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 an explicit conditional shortcut ("Pass outstanding=true for only the unpaid ones — that is the 'who owes us money' question") and enumerates the alternative filter axes (direction, docType, status, paymentState, currency, customerId, search). It also states the negative case directly: issuing, sending, cancelling or marking paid are financial acts "not available here", so the agent knows not to look for a write path on this tool.

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