update_draft

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

With no annotations, the description carries the full burden and does well: it discloses patch semantics ("Only the fields passed change"), list-replacement semantics for recipients, a destructive gotcha (body replaces the entire body including the quoted thread), a format contract (same Markdown as create_draft, html=true for ready-made HTML), and validation behavior (refuses non-drafts). It stops short of saying what is returned or how failures surface (bad draft_id, permission 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?

Two short paragraphs, front-loaded with the core purpose and mutation semantics before the body-format caveat. A couple of sentences are dense rather than wasteful, and the lowercase sentence opening with "body" reads as a formatting slip rather than a real cost.

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 six-parameter mutation tool with no annotations and no output schema, the description covers the semantics an agent needs to call it safely — partial updates, list replacement, whole-body overwrite, and content format. What is missing is the response shape and error behavior, which matters somewhat more given there is no output schema to lean on.

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%, so the description must compensate, and it largely does: it maps subject, body, to and cc to meanings, explains the absent-field-is-unchanged rule, states that recipient lists are replaced wholesale, and ties html to body formatting. Gaps remain — it never distinguishes to from cc, never clarifies how null differs from omission, and never names draft_id, though that parameter is self-evident from the required field.

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?

"Change an existing draft's subject, body or recipients" gives a specific verb (change), a specific resource (existing draft) and the mutable fields. "Refuses anything that is not a draft" further constrains the target, so an agent can distinguish this from create_draft, create_reply_draft and send_draft without opening any 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?

The description scopes usage to "edits the user asks for before sending", which is a clear when-to-use signal tied to the draft lifecycle. It also supplies an explicit when-not: for reply/forward drafts, do not edit the body, "re-create those drafts instead" — this routes the agent away from the tool toward creation siblings, though it does not name them directly.

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