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

With annotations only declaring readOnlyHint=false, openWorldHint=true, and destructiveHint=false, the description adds rich behavioral context well beyond them. It discloses the preview dry-run with no creation or email, fee mechanics (Stripe takes no Corply fee vs Corply Pay's 1% software fee), card vs ACH clearing times, idempotencyKey reuse on retries, AUTHORIZATION_CHANGED on mismatched requests, and the security constraint never to accept card or bank details in chat. It also describes what the confirmed call returns (payUrl, pdfUrl, emailDelivery) and how queued differs from received. No annotation contradiction exists.

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 description is front-loaded with the core purpose, which is good, but it is very long and includes some redundancy and generic boilerplate. The confirmation requirement is stated twice—once in the flow description and again in the closing 'Confirmation boundary' section. The 'Canonicality' and 'Idempotency' closing sentences are generic meta-guidance that is only partially relevant and repeats the earlier idempotency instruction. These passages dilute the otherwise information-dense content.

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?

Given a high-stakes mutation tool with no output schema and only sparse parameter descriptions, the description is largely complete. It covers the preview/confirm workflow, fee semantics, supported payment methods, error recovery, security boundaries, and returned values. The main gaps are the undocumented parameters noted above, but the guidance is sufficient for an agent to call the tool correctly as long as the schema fills in the remaining parameter-level details.

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 very low at 29% across 14 parameters, so the description must compensate. It does add substantial meaning for the core financial parameters: it explains the lineItems total as quantity x unitAmountCents plus stated taxes, details rateBps (8.25% -> 825) and fixed amountCents, and clarifies previewSha256/sendConfirmed usage and idempotencyKey reuse. However, it leaves several parameters unexplained, including companyId, payerName, payerOrganization, memo, title, allowedMethods, and _corply_context, so gaps remain.

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 does not state what the tool does. A description should describe what the tool does. A tool description must describe the tool's purpose. Write a useful description. The tool emails an itemized invoice the payer can pay by card or US bank account. This is a specific verb and resource. This is not a tautology of the name or title. It distinguishes from sibling payment tools like request_payment by specifying invoice, email delivery, and Corply Pay link. An agent can identify the tool's function without opening any 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?

The description provides clear procedural context: preview first without previewSha256, then confirm with a fresh plain-text confirmation and previewSha256. It gives explicit error handling routing to set_up_corply_pay for CORPLY_PAY_NOT_READY or CORPLY_PAY_TERMS_REQUIRED, and instructs not to call the confirming step when the invoice card's Send button is used. However, it does not explicitly compare send_invoice to sibling tools like request_payment, request_money, or manage_payment_request, so no exclusion guidance for those alternatives is present.

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