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

Annotations only supply destructive/openWorld/idempotent flags; the description goes well beyond them, disclosing required verified sending domain, inheriting suppression/tracking/sender-name/rate/storage protections, requestId reuse semantics for the same approved message, and that 'the response confirms provider submission, not delivery.' It also prescribes recovery behavior for stored=false and uncertain outcomes, which the annotations cannot express.

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?

Ten dense sentences, but the operational constraint (must show the message and get explicit authorization) is front-loaded immediately after the purpose statement. Nearly every sentence carries safety-critical behavior; a small amount of restatement around authorization could be trimmed, so not a full 5.

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?

No output schema exists, and the description does compensate by explaining what the response does and does not mean (provider submission, stored flag). For a destructive, open-world send tool it is unusually complete, though it leaves the draft/reply linking parameters (draftId, expectedDraft, inReplyTo, replyId) and locale/trackOpens unexplained.

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 description coverage is only 17% across 12 parameters, so the description must carry weight. It adds real meaning for `to` (1–5 explicit recipients), `from` (authenticated business mailbox), and `requestId` (stable UUID, reuse, never change content/recipients), but says nothing about draftId, expectedDraft, replyId, inReplyTo, locale, or trackOpens. Partial compensation merits only a baseline 3.

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+resource+scope: 'Send the exact approved personal email from the authenticated business mailbox to 1–5 explicit recipients.' It also explicitly demarcates itself from siblings by declaring 'This is personal correspondence, not a newsletter or bulk broadcast,' ruling out send_broadcast and send_customer_welcome. An agent can route to it 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?

Gives explicit preconditions ('First show the sender, recipient addresses, subject and full body; send only when the user explicitly authorizes that message'), explicit non-authorizing conditions ('Selecting CRM leads or preparing meeting/booking links never authorizes sending'), and names the substitute source for recipients (get_business_mail_recipients). This is a full when/when-not/alternative specification.

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