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

Annotations already declare destructiveHint=true and idempotentHint=true, but the description adds real value beyond them: token expiry, invalidation on record changes, and the required approval capability. It stops short of saying what rejection does downstream or whether it is reversible, which is the one gap for a destructive mutation.

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 critical constraint (apply only the exact previewed change, after confirmation) is correctly front-loaded. But the third sentence about an optional reason and note is dead weight and actively harmful because it does not match the tool's actual signature.

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, open-world mutation with no output schema, the description covers the workflow, prerequisites and token lifecycle well. It omits what the rejection actually changes, whether it can be undone, and gives a conflicting picture of the accepted inputs, so it is only partially complete.

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?

The description does add meaning for confirmation_token by tying it to the preview output and requiring a fresh one when stale. However, 'using an optional supported reason and note' advertises inputs that the schema forbids (additionalProperties=false, single required property), which will mislead an agent into trying to pass reason or note.

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 sentence gives a specific verb and resource and names the sibling it depends on (preview_reject_invoice), so an agent can tell it apart from the preview tool. The follow-up 'Prepare rejection of one invoice...' sentence muddies the picture, since 'prepare' describes the preview step rather than the apply step this tool performs.

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 states explicitly when to call it (only after the user confirms the preview), what it accepts (the confirmation_token only), and what invalidates a call (changed or expired records require a new preview). It also names the prerequisite capability (confirmation and approval), leaving nothing to inference.

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