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

Annotations declare readOnlyHint=false and destructiveHint=false, but the description does not mention that save_to overwrites an existing file — that destructive side effect is disclosed only in the input schema, and nothing in the description acknowledges it. On the other hand, the description does add genuinely behavioral context the annotations cannot convey: it blocks until the conversion finishes, and it bills one BankConv page per PDF page. The pagination cost is useful, but the gap on the overwrite risk against a false destructiveHint keeps this at 3.

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?

It is front-loaded with purpose and sibling routing before the output-mode list, and there is no filler. It is a long, dense block of comma-spliced clauses, but each sentence carries information.

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, so the description must explain return values — and it does, naming the returned fields and the saved-file path plus warnings for file outputs. But it never says what the returned 'summary' contains, and it omits the overwrite behavior of save_to. Good coverage of errors and timing, with a real gap in the default-return shape.

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 description's recap of output modes largely restates what the enum description already says. What it does add — that output="transactions" returns structured JSON, that every other value saves a file — is real semantics, but the save_to, password and timeout_seconds parameters are left entirely to the schema. With the schema doing the heavy lifting, a 3 is fair.

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 ('Convert a PDF bank or credit card statement on this computer with BankConv') and immediately distinguishes itself from the siblings that handle already-converted statements.

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 says to use this tool for a new PDF and to use get_transactions or download_statement with a conversion id for one converted earlier, naming list_conversions as the way to find that id. The when-not path and the alternatives are both spelled out.

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 past the annotations: it discloses the return shape (saved path plus warnings, NOT file contents), that no pages/credits are consumed, the 24-hour data-retention expiry, and four named error codes with the condition that triggers each. That is exactly the mutation/extension context the readOnlyHint=false / openWorldHint=true annotations leave unstated.

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 core action and id source, then return shape, then formats, then errors — a sensible order. It is somewhat dense and duplicates the format enum strings already present in the schema, which is the one wasted block.

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 steps in to describe the return value (path plus warnings, not contents) and enumerates the failure modes an agent must handle. For a 4-param file-writing tool that is sufficient to invoke and interpret 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 four parameters, including the format enum and save_to's overwrite behavior. The description's format list is a near-verbatim restatement of the schema enum text and adds no syntax or defaults beyond it — baseline 3.

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?

Opening clause states a specific verb and resource — 'Save an earlier BankConv conversion as a file on this computer' — and immediately disambiguates from convert_statement (which produces the conversion) and list_conversions (which supplies ids). An agent can tell exactly which sibling this is not.

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?

It clearly frames this as a second step ('an earlier conversion', id from convert_statement or list_conversions), so the workflow position is explicit. It stops short of stating exclusions, e.g. that this cannot reproduce an expired conversion except via the 'gone' error note.

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, so the trailing 'Read-only' is redundant with structured data. The description does add one genuinely useful behavioral fact — that each PDF page consumes one credit and the balance is scoped to the current period — but says nothing about what happens when the balance hits zero, when the period resets, or the response shape.

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?

Three short sentences, each carrying distinct information (what it returns, how credits are consumed, why it matters). The purpose is front-loaded and there is no filler beyond the redundant 'Read-only'.

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 zero-parameter read with no output schema, the description covers what is being measured, the units, and the reason to call it. Only the JSON shape and exhausted-quota behavior are left implicit, which is minor.

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?

The tool takes zero parameters, so there is nothing to disambiguate; baseline 4 applies. The description cannot and need not add parameter detail.

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 remaining BankConv pages), the quota unit (one PDF page = one page), and the period scope (current period). It is clearly distinguishable from the sibling conversion/download tools, though it does not name them explicitly.

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 usage trigger: check before converting a large batch, which implicitly routes the agent from convert_statement to this tool first. There are no explicit exclusions or named alternatives, but the decision context is clear.

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 flag readOnlyHint and openWorldHint, but the description adds real value beyond them: no page consumption, the 24-hour retention window that makes a conversion 'gone', and four named error codes including insufficient_credits. This is strong disclosure; only minor gap is no note on result size or ordering guarantees.

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?

Purpose is front-loaded in the first clause, followed by return shape, the alternative, and errors. The error enumeration runs long in a single sentence, but every item is actionable, so little is wasted.

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 must describe the return payload, and it does: date, description, signed amount, running balance, plus bank, account, period and currency. Combined with error codes and retention behavior, an agent has enough to call and interpret results; only edge cases like empty conversions or pagination are unaddressed (and pagination is stated as not applicable).

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 both parameters are already documented, and the description only repeats the conversion_id provenance. It adds nothing about the advanced 'tables' parameter's page:table syntax or when to override the default.

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 (get the transactions of a conversion) plus the identifier used to locate it, and names the sibling tools that produce that id (convert_statement or list_conversions). An agent can distinguish this from list_conversions and download_statement without opening either 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?

Explicitly says where conversion_id comes from and names the alternative for a different goal: 'To save a file instead, use download_statement.' That is a clear positive routing instruction with no ambiguity.

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?

Beyond the annotations it discloses ordering (newest first), the one-line-per-item return shape and the specific fields returned, which is real value since there is no output schema. 'Read-only' merely restates readOnlyHint and the 'recent' window is never quantified, so it falls just short of 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?

Two sentences, front-loaded with the verb and resource, then the use cases. Every clause carries information; nothing is redundant except the short 'Read-only' tag.

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 usefully spells out the returned fields and ordering, which covers the main completeness gap. Minor omissions remain: how 'recent' is bounded and what happens past the 100-item cap, though the limit parameter largely addresses the latter.

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% with both parameters fully described (including the enum semantics for status), so the schema does the heavy lifting. The description does not mention limit or status filtering at all, leaving no added meaning beyond the structured fields — baseline 3.

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 ('List the BankConv account's recent conversions') plus sorting ('newest first') and per-item fields returned. That scope makes it clearly distinct from convert_statement, get_transactions and download_statement without needing to open any 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 two concrete use cases — find a conversion id to feed get_transactions or download_statement, or check on a still-processing conversion — and names the sibling tools it precedes. The agent knows exactly when to reach for it rather than a sibling.

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