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

Beyond the annotations (readOnlyHint=false, idempotent, non-destructive), the description discloses the critical downstream boundary: nothing is sent or emailed, sending remains manual, no contractor dispatch, no purchasing-approval change. It also explains the two-mode outcome (durable text vs. proposal for manager/admin review), which is genuinely decision-relevant context.

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 the usage trigger and compact overall, but the non-sending point is restated three times ("never sends the message", "Sending remains manual", "Never emails a supplier"), while parameter meaning is omitted — space is spent on emphasis rather than on information the agent lacks.

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 7-parameter write-shaped draft tool with no schema descriptions and no output schema, behavior is well covered but input semantics are not. An agent could still supply garbage for rationale or mismatched expectedApprovalRequired, so the definition is only partially complete for correct invocation.

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 0% across 7 parameters, so the description carries the full burden, yet it only indirectly implies vendorId (active vendor), workOrderId (authorized work order), and subject/body. It says nothing about the required `rationale`, the `idempotencyKey`, or `expectedApprovalRequired` — the last being especially important since it interacts with the approval-preference logic the description does describe.

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 (prepare/draft) and resource (vendor message text) with clear scope: a vendor linked to an authorized work order. It doesn't explicitly contrast itself with the numerous sibling draft_* tools, but the vendor-communication focus makes the resource unambiguous.

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?

"Use this when preparing a message for a vendor linked to a job" gives an explicit trigger, and the description explains conditional behavior (prepare immediately if the vendor-message approval preference is off, otherwise produce a proposal for review). It stops short of naming alternatives or stating when not to use it, e.g. versus draft_maintenance_note.

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