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

Annotations only supply destructiveHint=false, so the description carries the real behavioral load and delivers: no delivery occurs and review/send happens in Gmail, the required gmail.compose scope with a reconnect hint for older accounts, the Reply All construction (original sender + To/Cc) and threading headers, and the fact that the result reports account, to and cc. This is far beyond what the annotations convey.

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?

A single dense paragraph that is front-loaded with the core verb/resource before the conditional details. Every clause conveys distinct information (mode, scope, reply semantics, account routing), though the parentheticals and em-dash asides make it slightly heavy for a description field.

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?

Covers the auth requirement (gmail.compose), multi-account routing, reply vs. new-mail behavior, threading, and even the shape of the result despite there being no output schema. For an 8-parameter mutation tool with no annotations of substance, nothing an agent needs to invoke it correctly is missing.

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%, so baseline would be 3, but the description adds meaning the schema does not: it explains that thread replies are drafted in the receiving account with Reply All semantics and proper threading headers, that reply_mode governs automatic recipient selection, and that omitting `to`/`subject` on a reply triggers Reply All and 'Re: <subject>'. These are cross-parameter behaviors the per-field schema text cannot express.

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 and resource ('Create a Gmail DRAFT (saved, not sent)') and immediately scopes the two modes: new mail vs. reply when thread_id is set. It explicitly distinguishes itself from the sibling send_message, so an agent can tell them apart without opening either 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 routing conditions: use thread_id (from search/get_thread) for replies, reply_mode='sender_only' only when the user explicitly asks to reply to the sender only, and pass `account` when more than one Gmail account is connected. It also states the no-alternative case up front ('Unlike send_message this never delivers anything, so no pre-send confirmation is needed').

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