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

Annotations declare readOnlyHint=false and destructiveHint=false, so the description doesn't need to restate safety. The description adds valuable behavioral context: it creates a draft (not sending), routes automatically by provider, and requires user confirmation before sending. It also discloses that the draft is created in the connected mailbox. The only minor gap is not detailing what happens on failure or whether drafts are saved to drafts folder, but the core behavior is well covered.

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 description is compact and front-loaded with the core purpose. The summary section is two sentences, and the returns section is one line. It avoids redundancy with the schema. The only slight inefficiency is the parenthetical provider list, but it's useful context. Overall it earns its place without bloat.

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 6-parameter tool with 100% schema coverage and no output schema, the description covers the essential context: what it does, provider routing, confirmation requirement, and email-finding alternatives. The returns section describes the output shape (success, draft_id, provider, draft details). The only missing piece is explicit error-handling or edge-case behavior (e.g., invalid recipient), but the schema's detailed parameter docs compensate. This is complete enough for an agent to invoke correctly.

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 100%, so the schema already documents all parameters thoroughly. The description adds value by explaining the 'to' parameter must always be a list of dicts, and the body parameter's Markdown rendering behavior. The schema itself is exceptionally detailed (including examples and rationale for optional name), so the description doesn't need to repeat that. The description's mention of provider routing and confirmation flow complements the schema's parameter docs.

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 description clearly states the tool creates a new email draft in the user's connected email account, names the supported providers (Gmail, Outlook, Superhuman), and notes automatic routing. It distinguishes itself from send_email by explicitly instructing to ask for confirmation before calling send_email, and from draft_reply by focusing on new drafts rather than replies. The verb 'create' and resource 'email draft' are specific and 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?

The description provides explicit when-to-use guidance: it tells the agent to always present draft details and ask for confirmation before calling send_email, and it names alternatives for finding email addresses (search_emails, search_linkedin_message_history, find_email). This is strong routing guidance that helps the agent decide when to use this tool versus siblings.

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