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

Annotations declare readOnlyHint=false, openWorldHint=true, and non-idempotent, so the agent knows this is not a pure read. The description compensates by disclosing the return shape (one-line verdict plus every problem with the invoice's own numbers) and by implying a job is created that get_result can read. It does not explain the cost/latency or side effects of that job, keeping it below 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?

Front-loaded with the checks, then the return value, then input forms, then the sibling alternative. Every sentence carries distinct information with no filler.

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, and the description fully covers the return value ('a one-line verdict and every problem found, with the invoice's own numbers'), plus input forms and follow-up routing. Nothing an agent needs to call it correctly is missing.

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 100%, so all four file parameters are already documented in the schema. The description's 'file_url, file_base64 or (ChatGPT) file' restates rather than extends that guidance, so the baseline of 3 applies.

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 ('Check whether an invoice adds up') and enumerates exactly what is verified: line quantity × unit price, lines vs subtotal, tax vs printed rate, subtotal + tax vs total. This clearly separates it from extract_invoice_data, invoice_to_excel, and get_result.

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?

Explicitly routes the agent for follow-up work: 'For additional outputs from the same processed invoice, use get_result with its job_id instead of submitting it again.' It also explains accepted input forms per client. It does not explicitly state when to choose this over extract_invoice_data, but the stated purpose makes that boundary obvious.

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