Match PO, goods receipts and supplier invoice

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

Annotations declare readOnlyHint=false and idempotentHint=false, so the agent needs to know what the write actually is; the description only offers 'save optionally' without saying what gets persisted or where. It does add genuinely useful behavior beyond annotations — the 500-line comparison cap, the return of source evidence and exceptions, and that offsetting errors are surfaced.

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 with no filler; the comparison scope is front-loaded, followed by return behavior. Only 'All diagnosis is free' leans toward promotional phrasing rather than operational detail.

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 deeply nested, 5-parameter tool with no output schema and 20% schema coverage, the description leaves major questions unanswered: what save=true creates, how currency/supplier/reference mismatches are handled, and what 'offsetting errors' means as a matching rule. It gives a useful sketch of inputs and outputs but is not complete enough for confident invocation.

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 only 20%, so the description must compensate, and it does partially: it names the inputs conceptually (PO lines, accepted receipts, prior billed quantities) and adds the matching rule that comparison is by exact SKU on accepted quantities. It adds no format or structural guidance for po_reference, supplier, currency, totals, or the save flag, leaving significant gaps.

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 (Compare) over specific resources (PO/invoice lines, accepted receipts, prior billed quantities) and names the matching key (exact SKU). An agent knows exactly what the tool computes, though the sibling report_* tools are unrelated so no differentiation is needed.

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?

The line 'All diagnosis is free; save optionally' implies a diagnose-then-optionally-persist workflow and nudges the agent toward the default non-saving call, but it never states when saving is appropriate or what conditions warrant it. Usage is implied rather than prescribed.

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