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

Annotations declare non-read-only, non-idempotent, open-world, non-destructive; the description goes beyond them by disclosing the full return shape, including per-field confidence, and implying a job_id is created for later retrieval. It doesn't state cost, processing time, page limits, or whether the job_id is returned in the response.

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 sentences, front-loaded with the output content that matters most given there is no output schema. The long field enumeration earns its place because it substitutes for a missing return schema.

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 fully compensates by listing returned fields and confidence, and it explains input-mode selection. It is slightly incomplete on operational behavior: no mention of size/page limits, accepted formats beyond 'PDF or image', or how the job_id is surfaced.

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 baseline is 3, but the description adds selection logic not in the schema: use file_url, file_base64, or (ChatGPT) file, with the local-file/ChatGPT distinction. This meaningfully helps the agent choose among the mutually exclusive input modes.

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 ('Extract') and resource ('an invoice's data as structured JSON') and then enumerates the exact fields returned (vendor, totals, tax lines, line items, confidence). An agent knows precisely what this tool produces, unlike the sibling check_invoice which only validates.

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 to get_result for additional outputs from an already-processed invoice instead of resubmitting, which is valuable reuse guidance. It also tells the agent which file input form to use per client. It does not, however, say when to prefer this over check_invoice or invoice_to_excel.

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