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

Annotations only declare the generic profile (readOnly=false, openWorld=true, destructive=false), while the description adds concrete behavior well beyond that: it consumes the user's API key and monthly page quota, masks account numbers, stores nothing, deletes results after one hour, and expires the download links. It also notes the password is used once in memory and never logged – exactly the kind of mutation-adjacent disclosure an agent needs.

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?

Everything is in one dense, front-loaded paragraph that leads with what the tool does, then inputs, then returns, then privacy/quota. The pricing-and-signup URL is slightly promotional but carries practical auth/quota information, so it roughly earns its place.

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 enumerates the return payload (bank, period, totals, check status, flagged rows, first N transactions, download links) and their one-hour lifetime. For a 7-parameter, open-world tool that is complete enough to call 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 semantics: the choice between file_url and file_base64, and that statement_id is used 'instead of a file' to retrieve a still-converting job's result, which explains the wait_seconds polling loop. That is meaning beyond the field-level 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?

A specific verb and resource ('convert a bank or credit card statement... into transactions') plus supported input formats and the balance-check behavior. It does not, however, distinguish itself from the sibling check_transactions, even though 'every row checked against the balances' overlaps with what that name implies.

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 does state the input alternatives ('Give file_url or file_base64', and statement_id to fetch an earlier result), which is genuine usage guidance. But it never says when to call this versus check_transactions or combine_statements, nor what prerequisites apply beyond needing a DocFig key.

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