Update a draft

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

The description discloses key behavioral traits beyond the annotations: that threading headers are preserved, the new version is saved before the old one is binned, and that omitting fields retains existing values. This adds meaningful context about what is destroyed (old version) and the order of operations, complementing the destructiveHint and idempotentHint annotations. It does not contradict any annotations, and it provides useful operational details without being overly verbose.

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, each contributing distinct information: the core action, a key behavioral guarantee (carryover of threading headers, save-before-bin), and the update semantics (omission behavior). It is front-loaded with the main verb and resource, has no redundant words, and is efficiently structured for an AI agent to parse quickly.

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 mutation tool with 10 parameters and no output schema, the description covers the critical behavioral aspects: replacement, partial update, and destruction of the old version. It also integrates well with the annotations and existing schema descriptions for parameters like attach, folder, dry_run, uidvalidity, and from_address. Minor gaps exist (e.g., what happens if the draft does not exist, or how attachments are handled in the replacement), but the description is sufficiently complete for an agent to understand the tool's core function and side effects.

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 50%, with some parameters (uid, to, cc, body, subject) lacking descriptions. The description helps by explaining the partial-update semantics for all fields ('Omit a field to keep what the draft already had'), which clarifies how parameters behave. However, it does not elaborate on individual parameter meanings or constraints (e.g., format of uid, what 'to' expects). The description adds moderate value but does not fully compensate for the missing parameter descriptions.

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's action with a specific verb ('Replace') and resource ('a draft's contents'). It distinguishes itself from sibling tools like create_draft (creates a new draft) and delete_draft (removes a draft) by focusing on updating an existing draft's content. The mention of 'Threading headers are carried over' further specifies a unique aspect of this tool, making its purpose unambiguous.

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 implies when to use this tool (when you need to update an existing draft rather than create or send one) but does not explicitly state when to use it versus alternatives. It provides one usage guideline—'Omit a field to keep what the draft already had'—but lacks explicit exclusions or alternative tool references. Sibling tool names provide context, but the description itself gives no direct comparison guidance.

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