get_changes

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

Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false and openWorldHint=false, so the safety profile is covered. The description adds genuinely new behavioral context beyond the annotations: the payload carries evidence dates, conditions, coverage gaps and warnings, and absent context must not be treated as proof of eligibility. That is useful, but it says nothing about pagination, cursoring, or how partial results should be interpreted across pages.

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 the read-only property front-loaded, so it is appropriately sized and makes no redundant claims. However, the middle and last sentences are cryptic, especially 'Preserve evidence dates, conditions, coverage gaps and warnings,' which reads as an instruction to the reader rather than a description of the tool's behavior.

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?

An output schema exists, so return values needn't be explained, but with 7 undocumented parameters and no explanation of what a 'change' is or what from/to delimit, the definition leaves too much to inference. The one completeness strength is the warning that gaps in context do not establish eligibility, which partially addresses the domain risk of over-reading results.

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 7 parameters, so the description carries the full burden and it explains none of them: from/to semantics, the 1-50 limit, the cursor, the include='evidence' constant, and the release/programme filters are all undiscussed. The only indirect signal is the mention of 'evidence,' which never states that include must be 'evidence'. The gap is substantial for a filtered, paginated query.

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 opens with 'Read-only get changes,' which largely restates the tool name and never says what kind of change objects these are (travel-rule changes, programme changes, evidence snapshots) or over what domain 'from'/'to' range. It reads as a tautology with a hint of read-only scope rather than a specific verb+resource definition, and it does nothing to distinguish it from siblings like get_travel_rule or get_coverage.

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?

There is no when-to-use or when-not-to-use guidance, and no sibling is named among list_programmes, get_programme, compare_programmes, get_programme_costs, get_travel_rule, get_source, get_coverage. The clause 'Missing context never establishes eligibility or a travel waiver' is a downstream interpretive rule, not a routing rule, so the agent still has to guess which tool answers a change-history question.

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