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

Annotations already declare destructiveHint=true and idempotentHint=false, so the safety profile is partly structured; the description still adds real value beyond it by disclosing the auth scope, the fact that the first call mutates nothing, and that unknown fields are rejected by name. It does not say whether the resulting change is reversible or what happens if confirm_token expires, which matters for a destructive, non-idempotent operation.

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?

Front-loaded with purpose and scope, then the two-step protocol in short clauses with no filler; the terse fragments ("Two calls.") are deliberate and readable. A little clarity is sacrificed for brevity around the phrase "checks the arguments as the change would".

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 destructive mutation with no output schema and low schema coverage, the description covers the confirmation workflow and permissions but omits what is actually editable, what the returned summary describes, and failure/expiry behavior. It is enough to avoid blind mutation but not enough to call it confidently on the first attempt.

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 only 33%, so the description must compensate. It does explain confirm_token well (obtain from the first response, pass back with identical arguments) and hints that unsupported fields are rejected, but proposal_id and document_id are never explained — particularly proposal_id, which leaves the agent guessing whether a proposal must exist beforehand and how to obtain one.

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?

"Change a document" names a specific verb and resource and, by implication, separates it from add_document/remove_document/convert_document in the sibling list. It never says which aspects of the document can be changed or what proposal_id represents, so an agent knows the category of the operation but not its actual scope.

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?

It gives unusually concrete invocation guidance: two calls, the first is a dry run, show the summary to the person, then re-call with the same arguments plus confirm_token. It also states the required scope, bookings:write. What is missing is any when-not guidance or routing to a sibling (e.g. when to use this versus add_document or convert_document).

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