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

Annotations cover safety (readOnlyHint, openWorldHint), and the description adds genuinely behavioral context: historical coverage start date, the tri-state status (valid/invalid/unknown), the inclusive effective period, and base-unit normalization (1 or 100) that affects interpretation. Combined with source URL and hash disclosure this exceeds what annotations provide, though it remains silent on rate limits or caching.

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?

The definition is front-loaded with the core question, then input, then return shape, then caveats, so an agent can stop reading at any line and still act. It is dense and every clause carries information, though the enumeration of retrieval/verification timestamps and SHA-256 hashes is more output detail than the caller strictly needs given an output schema exists.

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-parameter read-only lookup with an output schema present, the description covers the key gaps an agent needs: the date floor, the unknown-result semantics, and the non-authoritative nature of the rate. Return-field detail duplicates the output schema, but nothing essential to correct invocation is missing.

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 meaning beyond the schema by specifying the inputs as an ISO 4217 code plus a YYYY-MM-DD Japan-calendar date and, via the earlier revision note, normalizing lowercase input to uppercase. That normalization behavior is not expressed in the schema properties themselves, which justifies a step above baseline.

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?

The opening question states a precise verb and resource: retrieve the JPY rate Japan Customs published for a given currency and date. It also scopes the value ('Reference for customs and bookkeeping only; not a market or remittance rate'), which distinguishes it from generic FX tools. However, it never differentiates itself from the sibling 'jp_customs_fx_batch', which an agent could easily confuse for bulk lookups.

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 context is clear: it states the two required inputs, that coverage begins 2021-08-15, and that unpublished periods or unlisted currencies return 'unknown' rather than an error. The exclusion of tax calculation and market/remittance use further bounds scope. It stops short of naming the batch sibling as the alternative for multiple lookups, so routing guidance is incomplete.

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