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

With no annotations, the description carries the full burden and does substantial work: it discloses non-retention ("The file is not stored"), a per-document price with a page cap (€0.05, up to 10 pages), billing tied to the API key in the Authorization header, and deterministic cross-field checks (qty × price = line amount, subtotal + VAT = total, delivery after order date, EU VAT-ID format). Not covered: failure behavior on non-PO documents, timeouts, or max file size.

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 input→output mapping in a single tight line, then validation, commercial terms, and parameters. The long field enumeration earns its place by defining the extraction contract, though the closing "Parameters —" block is written in a compressed register that slightly hurts scannability.

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?

An output schema exists, so return values need not be re-explained, and the description still summarizes the payload shape. For a two-parameter, stateless extraction tool it covers inputs, OCR defaults, validation guarantees, pricing, data handling and downstream compatibility — everything an agent needs before calling.

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 0% and the schema is bare (both fields nullable, none required), so the description must compensate — it defines url as the PO PDF or image URL and language as the OCR language set for scans with a concrete default list (en,de,fr,es,it,nl,pl,uk). It does not clarify that url is effectively mandatory despite the nullable schema, leaving a small semantic gap.

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 with an explicit mapping syntax: "Purchase order PDF, scan or photo → PO number and date, buyer, supplier, ship-to, delivery date, payment terms, Incoterms and every order line". This distinguishes it from generic siblings such as ocr, pdf_text, table_extract and invoice_parse by naming PO-specific fields and line-item granularity.

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 clear usage context: input must be a PO as PDF, scan or photo, OCR languages apply to scans, and the result is "Ready to feed invoice-parse mode=pro po= matching", which tells the agent where the output belongs in a workflow. It stops short of explicit exclusions (e.g. use invoice_parse for invoices, ocr for non-PO scans), so the agent must infer the boundary from siblings.

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