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

Annotations already mark this read-only, idempotent, non-destructive and closed-world, and the description still adds real behavior: it returns a Transactions sheet with named columns plus a Balance check sheet and a CSV, and it reconciles opening + credits - debits = closing to the cent. The privacy note ('does not send data to a bank') and the input-format limits are exactly the kind of context annotations cannot carry.

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 use case, then input formats, then outputs, then exclusions — a logical order with no filler sentences. The long run of quoted trigger phrases is slightly verbose, though it serves routing rather than padding.

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 two-input, file-plus-text tool with an output schema, the description covers formats, outputs, and reconciliation behavior thoroughly. The one residual gap is that it never says one of file/statement_text is mandatory, which matters since the schema marks nothing required.

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 baseline is 3; the description earns an increment by explaining that input arrives either as an attached file or as pasted text, which tells the agent the two parameters are alternative input modes. It does not state precedence or that at least one must be supplied, which keeps it below a 5.

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 plus resources: converts MT940 (.sta, .mt940, .txt) and ISO 20022 camt.053/.052/.054 into an Excel workbook and CSV, and additionally verifies statement totals. The scope is precise enough that an agent could not confuse it with a generic file or finance tool.

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?

Opens with an explicit when-to-use trigger ('the user wants to convert an MT940 file to Excel or CSV') and lists the literal phrases that should route here ('mt940 to excel', 'camt 053 to excel'). It also names exclusions — no PDF/CSV statements, no transaction categorization, no data sent to a bank — so misfires are easy to avoid.

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