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

Annotations only give the generic readOnly=false / idempotent=false / destructive=false profile. The description adds real behavioral substance beyond them: the client receives a private link valid 30 days, they answer Yes/No per optimization with a reason on each No, and resending generates a fresh link — which concretely explains the non-idempotent side effect. It also flags the pre-send confirmation requirement.

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 tight sentences, front-loaded with the actor and action, then the client-side mechanics, then the safety caveat. Every sentence carries information; the only slight redundancy is that 'For an AGENCY' and 'Confirm with the agency' both restate the agency framing.

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 2-parameter, no-output-schema tool this covers channels, recipient, response mechanism, link lifetime, resend behavior and a pre-send check — enough for correct invocation. It stops short of 5 because it says nothing about delivery failure, invalid proposal ids, or permission requirements on the agency account.

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 50% and the 'via' array carries no per-item description, so the description must compensate — and it does, spelling out 'email, text (sms) or both,' which clarifies that the array accepts either or both enum values. The 'proposal' parameter's linkage to draft_optimization_proposal is only in the schema, but the prose reinforces the 'drafted' precondition rather than adding error/format details.

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 (send), resource (optimization proposal), actor (AGENCY) and target (the client) in the first clause, and names the sibling concepts it is not — drafting is covered by draft_optimization_proposal, and this tool is the delivery step. An agent can distinguish it from my_optimization_proposals or draft_optimization_proposal without opening any schema.

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?

Usage context is clear: the proposal must already be drafted (the parameter points at draft_optimization_proposal), and the description adds the human-in-the-loop guard 'Confirm with the agency before sending.' It does not name an explicit when-not case or a competing sibling to route against, so it stops short of full when/when-not guidance.

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