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

Annotations cover the safety profile (readOnlyHint=false, destructiveHint=false, idempotentHint=false), and the description adds substantial context beyond them: it replaces any earlier draft on the message, adds it to the to-do list, surfaces it under E-post › Utkast, returns stored draft text and replyArgs, and documents character limits and MAIL_NOT_FOUND/MAIL_DRAFT_TOO_LONG errors.

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?

Purpose and the 'sends NOTHING' caveat are front-loaded, and most sentences earn their place (side effects, error codes, send handoff). It is still a dense single block, and the send-path instructions could be tightened since they describe a different tool's invocation.

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 3-param write tool with no output schema and no param descriptions, the definition covers everything an agent needs: scope, side effects, replacement behavior, return payload, language rule, size limit, and the two failure codes — all before the agent calls it.

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?

With 0% schema description coverage the description must carry the load, and it does for two of three params: 'one received message' plus 'MAIL_NOT_FOUND means the id is not a received message' scopes the id, and the ~3,500-character cap constrains text. The locale parameter is only indirectly addressed via 'Write in the language of the conversation,' so its enum values remain undocumented.

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 precise verb+resource: 'Save a reply draft on one received message in the authenticated business mailbox, for the owner to review,' and explicitly clarifies it 'sends NOTHING.' This clearly separates it from the send path and from list/get siblings in the business-email family.

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 when-to-use (drafting for owner review), when-not (not authorization to send), and the exact alternative path: show the owner the draft, get explicit approval, then call send_business_email with replyArgs, new requestId and approved:true. It even names the sibling tool, leaving nothing to inference.

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