mail_save_draft

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

Annotations already declare a non-read-only, non-destructive, non-idempotent, open-world operation, so the safety profile is covered. The description adds real context beyond them — threading preserved on replies, attachments preserved on forwards, return of the new UID/UIDVALIDITY, and that it never sends. However, it omits permission requirements and whether repeated calls create duplicate drafts (relevant given idempotentHint=false).

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?

Four short sentences, no filler, with the core action and the conditional prerequisites front-loaded and the never-sends guarantee placed last for emphasis. Slightly dense but every sentence carries information.

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 helpfully states the returned UID and UIDVALIDITY, which closes that gap. For a 12-parameter mutation tool with zero schema descriptions, though, important call-shaping details (required accountId, address list limits, attachment relativePath semantics) are absent.

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 0% across 12 parameters, so the schema documents none of them. The description compensates for the conditional parameters — folder, uid, uidValidity for reply/forward, attachments for forward, recipients/body — but is silent on accountId, cc, bcc, subject, html/text distinction, kind default, and the attachment object shape, leaving much of the surface 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?

The description states a specific verb (save) and resource (server draft) and enumerates the supported kinds: new, reply, reply-all, forward. The closing 'Never sends' cleanly separates it from mail_send_prepared, though it never names mail_update_draft or mail_prepare_draft as alternatives.

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?

It states concrete preconditions — reply/forward requires the original folder, UID and observed UIDVALIDITY — and notes recipients and body may be incomplete, which tells the agent this can be called with partial data. It lacks any explicit routing advice for modifying an existing draft versus creating a new one.

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