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

With no annotations, the description carries the full burden and mostly succeeds: it discloses per-document pricing, that billing is charged to the API key in 'Authorization: Bearer', the default OCR language set, and the deterministic checks performed (lines vs subtotal, base × rate vs VAT, IBAN checksum, EU VAT-ID format, due date). The pricing statement is internally muddled — 'mode=pro (€0.06)' alongside 'Cost €0.03 per document' — which leaves the agent unsure of actual cost.

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 extraction target list is well front-loaded, but the body is a dense run-on with a duplicated 'Parameters —' restatement of mode that repeats the earlier mode=pro sentence. The repeated pricing figures differ (€0.06 vs €0.03), so words that should clarify instead create ambiguity.

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 explained; the description covers the remaining agent-critical facts: cost, auth header, defaults, mode differences and PO matching. Gaps remain around failure handling and the fact that url is optional in the schema despite being the core input, but overall it is sufficient to invoke correctly.

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 0%, yet all four parameters are explained with meaning and defaults: url (image or PDF location), language (OCR languages, default en,ru,uk,de), mode (basic default | pro), and po (pro-only purchase order JSON with an explicit {number, items:[{sku, description, qty, unit_price}]} shape). This adds far more than the bare anyOf string schema provides.

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?

Opens with a specific verb-implied resource list: 'Vendor, customer, number, dates, currency, totals, VAT and line items from a photo, scan or PDF invoice.' The input modality (photo/scan/PDF) and the cross-check behavior distinguish it from siblings like ocr, bank_statement_parse and e_invoice without needing to open the 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?

Gives a clear cost/feature tradeoff for mode selection: basic at €0.03 for totals-level extraction vs pro at €0.06 for SKU/per-line VAT, VAT breakdown by rate, references, checks and PO matching. It does not, however, route the agent away from overlapping siblings such as invoice_3way_match, purchase_order_extract or plain ocr, so tool-level exclusions are missing.

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