create_draft

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

Annotations are minimal (all hints false), so the description carries the burden. It adds that the tool returns a draft ID with status 'draft', which is helpful. However, it does not disclose behavioral traits like non-idempotency (multiple calls create multiple drafts) or any side effects beyond creation. The description does not contradict the 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?

The description is three sentences that front-load the core purpose, immediately follow with usage guidance, and end with return format. Every sentence earns its place with no redundancy or fluff. It is perfectly scoped for the tool's complexity.

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?

Given 5 parameters, no output schema, and moderate complexity, the description covers the main points: purpose, two usage modes, and return structure. It could mention that drafts are persisted per mailbox and that the tool is not idempotent, but these gaps are minor. The sibling tool set (list_drafts, get_draft, update_draft, send_draft) fills in some missing context, and the description is sufficiently complete for correct agent invocation.

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% with good descriptions for all parameters. The main description adds value by explaining how parameters interact: 'Same fields as send_email, or omit to/subject and pass reply_to_message_id to draft a threaded reply.' This provides pragmatic usage guidance beyond the schema's individual parameter descriptions, especially for the conditional omission of to and subject.

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 prepares an email draft without sending it, using the specific verb 'prepare' and the resource 'draft'. It explicitly distinguishes from the sibling send_draft and send_email, and also mentions an alternative usage for threaded replies via reply_to_message_id, making the purpose unmistakable.

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 gives explicit context: 'Call this when you want to stage a message for later or for a human to review before it goes out — send it afterward with send_draft.' This tells when to use the tool and points to the sibling for the next step. It also explains the alternative threading workflow. However, it doesn't explicitly state when not to use it (e.g., if immediate sending is needed, use send_email), but the context is strong enough for a high score.

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