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

Annotations declare it is a write (readOnlyHint=false, idempotentHint=false, destructiveHint=false). The description adds genuine behavioral context: requisites are validated by checksums, bank details are optional for an act, the tool writes a file to disk and needs no network, and it returns a file path, totals and the PDF. However, idempotency/non-destructiveness and the write nature are already carried by annotations, so the added value is useful but not exhaustive (no overwrite/conflict behavior).

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?

A single dense but front-loaded block: creation purpose first, contents next, usage and routing to the sibling after, then behavior and return values. Every sentence carries weight, though the run-on semicolon structure is slightly heavy.

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?

For a mutation tool with nested inputs and no output schema, the description covers the essentials: what is produced, when to use it, validation behavior, that bank details are optional, that it writes to disk, and what it returns. Nothing an agent needs to invoke 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% and every parameter, including nested seller/buyer and item fields, is documented in the schema, so the baseline is 3. The description adds only marginal semantics beyond that — that bank data is not required for the act and that items may be reused from invoice_create — without contributing format or validation detail the schema lacks.

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 (creates an act of completed works as PDF) and enumerates the document contents (executor, customer, basis, period, services table, VAT total, amount in words, signature blocks). It explicitly differentiates itself from its sibling by naming invoice_create as the tool that issues invoices instead.

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 an explicit when-to-use condition — 'закрывающий документ после оплаты или выполнения работ' — and names the alternative tool (invoice_create) with the note that the same line items can be reused. No inference is required to choose between them.

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

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

Annotations already declare readOnlyHint=true and openWorldHint=false; the description reinforces the latter with 'Только вычисление, без сети' (calculation only, no network) and discloses output behavior (kopecks as digits by default vs fully in words). It does not need rate-limit/auth detail for a pure offline computation, so 4 rather than 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?

The output format example and the primary purpose are front-loaded, followed by currency scope, usage, and the offline note. Every sentence carries distinct, non-redundant information and none could be removed without losing value.

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?

For a 3-param, no-output-schema computation tool, the description covers input semantics, output format, supported currencies, and routing versus siblings. An agent has everything it needs to call and interpret the tool 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 100%, so the baseline is 3, but the description adds real meaning: it explains currency=none for a plain number without currency and maps minor_in_words to the 'kopecks in digits vs words' behavior with a worked example. This goes beyond the schema's terse field descriptions.

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+resource (converts an amount into text) and immediately shows the exact output format, so the agent knows precisely what it produces. It also distinguishes itself from siblings by noting that invoice_create/act_create already embed this conversion, making it unambiguous versus peers.

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 explicit when-to-use (contracts, receipts, payment orders) and when-not-to-use (invoices/acts from invoice_create/act_create already contain the amount in words), naming the sibling alternatives directly. Nothing about tool selection is left to inference.

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

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

Annotations already declare readOnlyHint=false, openWorldHint=false, idempotentHint=false, destructiveHint=false, but the description goes further and adds genuinely useful behavior: it validates INN/KPP/BIK and BIK consistency and refuses to write the file on error, it writes the PDF to disk (side effect location), it needs no network, and it returns the path plus computed totals and the PDF itself.

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 purpose, then validation behavior, then sibling routing, then side effects — a sensible order with no filler sentences. It is dense, but every clause carries information; only the fine-grained enumeration of document parts is slightly over-detailed.

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?

For a 12-parameter, nested-object tool with no output schema, the description covers the missing pieces: what the return contains (path, sum/vat_amount/total/total_words, the PDF), the validation-and-abort behavior, and the on-disk side effect. An agent has everything needed to invoke it 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 description coverage is 100%, so the schema already documents all 12 parameters, including the VAT enum values and the note that QR requires seller bank details. The description's mentions of НДС/«Без НДС» and сумма прописью largely restate what the schema and the returned-totals list already convey, so it adds little beyond the baseline.

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 artifact (создаёт счёт на оплату в PDF по типовой форме) and enumerates the exact document contents: bank block, parties, line items, totals, VAT, amount in words, signatures, QR per ГОСТ Р 56042. It also explicitly separates itself from the closing document (act_create) and the standalone QR tool (payment_qr), so an agent can route without opening schemas.

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?

Names both alternatives and the condition that selects each: act_create for a closing document, payment_qr when only the QR is needed, invoice_create for the invoice itself. It also states the precondition (INN/KPP/BIK validation must pass, otherwise no file is produced).

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

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

Goes well beyond the annotations (readOnlyHint=false, destructiveHint=false, idempotentHint=true, openWorldHint=false) by disclosing that it writes a PNG to disk, needs no network, validates recipient details, and what it returns (image, file path, payload string). These are the exact traits an agent needs for a local file-writing tool.

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-loads the core capability and standard, then the routing hint and the side effects. Dense but every clause adds information; the laundry list of supported banks is the one slightly expendable element.

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 compensates by describing the return values (PNG image + file path + payload string) and the disk-write side effect. For a nested-object, 4-parameter tool, nothing essential 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%, including nested recipient fields, so the schema already carries parameter meaning. The description adds only the high-level notion that recipient/account/amount/purpose end up on the payment form; baseline 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 artifact (creates a payment QR per GOST R 56042 / ST00012), names the consuming apps, and explicitly distinguishes itself from invoice_create. An agent can identify the tool's output without opening 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 concrete use case (send in a messenger instead of bank details) and an explicit alternative with its selecting condition: 'Если нужен полноценный счёт с QR внутри — invoice_create.' The routing decision is fully resolvable from the description.

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