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

Annotations already declare readOnlyHint=true and destructiveHint=false, yet the description still adds genuinely new behavior: a five-minute token expiry, the preservation of correction/audit history, and an explicit no-mutation guarantee. The token TTL is operationally critical and appears nowhere in the structured fields.

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 sentences, no filler, with the action and the read-before-write precondition front-loaded and the non-mutation guarantee held for the close. Every clause carries distinct information.

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 one-parameter preview tool with no output schema and full annotation coverage, this covers purpose, safety, and return shape. The remaining gap is the handoff: nothing says which tool accepts the five-minute confirmation token or what happens if it expires.

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?

One nested parameter with 0% schema description coverage, so the schema alone gives only UUID formats for invoice_id and line_item_id. The description compensates partially by imposing a read-before-write constraint ('previously read') and demanding an exact match, but never explains what either identifier refers to or where it comes from.

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?

States a specific verb-and-resource pair ('Prepare removal of ... line item') and immediately scopes it as a dry-run ('it does not apply the change'), which separates it from the sibling remove_invoice_line_item without either schema being opened. 'Exact previously read line item' also signals the required object identity.

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 confirmation-and-token framing tells the agent this is the first half of a two-step mutation, and 'does not apply the change' sets the when-not boundary. It stops short of naming remove_invoice_line_item as the follow-up call, leaving the agent to infer where the token is consumed.

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