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

Annotations already cover the safety profile (readOnlyHint=true, idempotentHint=true, destructiveHint=false), and the description adds substantive behavioral context: verdict semantics ('not_applicable... never counts as a pass', 'conforms is true when no finding failed'), the deliberate cbpr+ omission with rationale, and the auditable source-string requirement. No contradiction with annotations; the read-only framing aligns perfectly.

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?

Long, but deliberately structured with labeled sections (USE WHEN, DO NOT USE, SCHEMES, VERDICTS, IMPORTANT, FREE, COST) and the core purpose front-loaded in the first sentence. The verdict-semantics and scheme-rationale sections are essential for correct invocation and interpretation; only the cost paragraph approaches borderline content, and it is clearly delimited.

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 a full output schema, a nested input object, and both required params at 100% schema coverage, the description covers everything an agent needs: when to use, when not to, scheme identities, verdict semantics, the cbpr+ gap, auditability requirements, and cost. Even the November 2026 regulatory timeline orients the agent on why the check matters.

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 schema documents every field, satisfying the baseline. The description still adds value by expanding the scheme enum with publishing bodies ('sps (Swiss Payment Standards, SIX), hvps_plus (HVPS+ / T2, ECB), fedwire (Federal Reserve)') and explaining the deliberate absence of a cbpr+ option. The address sub-fields need no repetition since the schema already describes them fully.

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 a structured ISO 20022 postal address against a payment rail's published address rules, rule by rule, each verdict citing the document it comes from.' The DO NOT USE clause ('not postal reality') sharply distinguishes it from address-verification tools and from the sibling validate_iban/validate_payment_reference pair.

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?

Provides an explicit USE WHEN condition ('assembling a payment instruction... with a creditor or debtor address, to learn whether the rail accepts it BEFORE submitting') and an explicit DO NOT USE exclusion ('to verify that a street or town EXISTS'). It also instructs the agent on what to do with the output ('relay each finding's source string'), which is actionable usage guidance beyond mere selection.

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