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

Annotations already declare readOnlyHint, idempotentHint and destructiveHint=false, so safety is covered; the description earns credit for adding what annotations cannot: daily refresh, credit cost of 1 per call, an unauthenticated preview capped at 5 rows and 20 calls/day, and the disclosure that operator comments are dropped at ingest and artwork is not served. It does not describe pagination behavior or result ordering, but the added operational context is substantial.

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?

Dense but front-loaded: dataset scope and geography first, then the returned field list, then filter/cost mechanics. The field enumeration is long but load-bearing given there is no output schema. Only the artwork/comment sentence is arguably marginal, and even that sets expectations about missing data.

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?

This is a 20-parameter dataset query, and the description compensates well on the output side by enumerating the returned fields in the absence of an output schema, plus coverage/refresh/cost context. The decisive gap is input-side: with zero schema descriptions, an agent cannot confidently construct scheme_type, rating_value, or business_type filters from this text alone.

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 0% across 20 parameters, so the description carries the full burden and largely fails it: only q is explained ('searches all text fields') plus a note that filters AND together. Nothing clarifies scheme_type values, rating_value accepted strings, new_rating_pending encoding, business_type vocabulary, postcode vs outward_code semantics, or the *_min/*_max pair conventions. Parameter names are largely self-describing, which prevents a 1, but an agent still has to guess at value formats for several key filters.

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 description names a specific resource and its exact provenance: businesses rated under the UK Food Hygiene Rating Scheme (England, Wales, NI; Scotland FHIS) from the official FSA open-data file, keyed by FHRSID. The verb 'query' is only implied via the name, but the enumerated returned fields make it unambiguous what you get back. It is clearly distinguishable from siblings like query_uk_schools or query_uk_planning by the domain, though it never names an alternative explicitly.

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?

'Filters combine with AND; q searches all text fields' gives real guidance on how to combine parameters, and the credit/preview note tells the caller what happens without a key. But there is no when-to-use vs when-not, no statement of what happens when zero filters are supplied, and no routing to sibling datasets. 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.