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

Annotations already signal destructiveness (destructiveHint: true) and non-idempotency. The description adds real-world context ('emails a real person from the user's address') and reinforces the irreversible, consequential nature by linking it to explicit user approval. This goes beyond the structured annotations.

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?

Two sentences, both high-value. The first states the purpose and the second states the critical precondition. No fluff, front-loaded with the action and the approval constraint.

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 destructive send action, the description covers the approval requirement, the real-world impact, and the parameter linkage. Since an output schema exists, return values are already documented. Nothing essential is missing for an agent to call this safely and 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 coverage is 100% and already includes descriptions for body and confirmed, including the requirement that confirmed must be true only after approval. The description reinforces this by saying the body must be exactly as approved, and the mention of draft_kind ('Set when the body came from inbox.draft_reply') adds useful cross-tool context not present in the schema.

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 states a specific verb (send) and resource (email reply via Gmail), and explicitly notes it emails a real person from the user's address, distinguishing it from drafting or applying suggestions. It clearly separates this from sibling tools like inbox.draft_reply.

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?

Explicit when-to-use guidance: 'Only call after the user has seen the exact message body and explicitly approved sending it.' This is a strong guardrail that prevents premature or unauthorized sends. It also implies when not to call (without approval).

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