What changed between two versions

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

Annotations already indicate readOnlyHint, idempotentHint, and destructiveHint false, so the safety profile is covered. The description adds behavioral context by specifying the data source ('from its changelog or GitHub releases') and the output format ('each tagged with its version'), which is not present in annotations. It also imparts the constraint that 'to defaults to the latest,' which is a subtle but important behavior for parameter handling.

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 description is concise and front-loaded, putting the purpose ('Before an upgrade') first, which immediately signals when to use the tool. Every sentence earns its place: the first clarifies the use case and content, the second adds the version tags and default behavior. No fluff or redundancy.

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?

Given the tool's complexity (5 parameters, output schema present, and multiple sibling tools), the description is largely complete. It covers the main purpose, the parameters' roles (from, to, defaults), and the output structure. However, it lacks explicit guidance on the 'prereleases' parameter and does not elaborate on the output schema's structure beyond mentioning tags, which could be inferred. Despite these minor gaps, the description is sufficient for an agent to make a correct call in most scenarios.

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 40%, and only ecosystem and from have descriptions in the schema. The description adds context for 'to' by stating it defaults to latest, and clarifies the meaning of 'from' as the current version, which partially compensates for the low schema coverage. However, it does not explain the 'prereleases' parameter or the 'package' parameter beyond the schema's basic type, leaving room for ambiguity.

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 clearly states the tool's purpose: to list breaking changes, removals, deprecations, security fixes, and raised requirements between two versions of a package. It specifies the resource (package) and the scope (between two versions), and the mention of 'changelog or GitHub releases' differentiates it from simply listing raw release notes. This distinguishes it from siblings like release_notes and search_changelog, which likely focus on a single version or searching within changelogs.

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 description opens with 'Before an upgrade,' which sets a clear context for when to use this tool: when planning an upgrade. It explicitly lists the types of changes covered, which helps clarify what information the tool provides. However, it does not explicitly state when NOT to use this tool or name alternative tools, leaving the agent to infer that release_notes or search_changelog might be used for other purposes, such as viewing full release notes or searching for specific terms.

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