Audit VAT treatment of vouchers

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

Annotations already declare readOnlyHint=true and destructiveHint=false, so 'Read-only' in the description adds nothing. The description does disclose what the sweep looks for, which is genuine behavioral context, but says nothing about how much work it performs, what happens when maxVouchers is hit, or whether results are truncated — all material for an openWorld sweep.

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 front-loaded sentence: the action and outcome come first, then the defect inventory. Every clause names a distinct detection category, so the length is earned rather than padding, though the list runs long enough that an agent must parse carefully.

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 carry return-value semantics, and it only does so indirectly by naming the problem categories. It never says what a finding looks like (per-voucher list? severity? counts?), nor what happens with an unbounded period sweep — enough missing for a complex, open-world audit tool to be a real gap.

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 every parameter (from, to, maxVouchers, includePositions) is already documented in the schema with formats, defaults and trade-offs. The description mentions none of them; baseline 3 applies since the structured data carries the full load.

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+resource ('Sweep vouchers and flag VAT problems') followed by an enumeration of the exact defect classes it detects (wrong taxRule for foreign suppliers, revenue rules on expense vouchers, rate/rule mismatches, arithmetic errors, supplier inconsistency). This is far more concrete than a generic 'audit' label. It does not, however, acknowledge the overlapping sibling sevdesk_reverse_charge_report, which an agent could easily confuse with this 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?

Usage is implied by the defect list — you would call it before filing VAT returns or when reconciling suppliers — but there is no explicit 'use when'/'do not use when' framing and no routing to alternatives such as sevdesk_reverse_charge_report or sevdesk_find_duplicates. The 'Read-only' note signals it is a safe inspection step, but the agent must infer the scenario.

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