gmail_create_reply_draft

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

Beyond the annotations, the description discloses important behavior: PII tokens are restored before saving, secrets remain redacted, it syncs to Apple Mail's Drafts via IMAP, and it returns a self-minted Message-ID and deep links. The 'Does NOT send' caveat explicitly clarifies side effects. This is genuinely helpful context that annotations alone do not provide.

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?

Three dense sentences deliver the purpose, required inputs, behavioral caveats, and return value hints without wasted words. The most important action verb is front-loaded, and every sentence contributes meaning.

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?

The tool is reasonably complex, but the description covers the core workflow: inputs from a prior read, token restoration, redaction persistence, threading, IMAP sync, return artifacts, and the no-send caveat. An output schema exists, so the return-value details are not strictly required in prose, and the description still summarizes them enough for an agent to know what to expect.

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?

The schema coverage signal is 0% for the top-level params object, so the description compensates by mapping the original message fields ('from', 'subject', 'threadId', 'message_id') to the required inputs and explaining the PII-token handling. It does not mention the optional 'references' parameter, and exact formats are left to the schema, so it is strong but not exhaustive.

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 opens with a specific verb+resource: 'Create a threaded REPLY draft to an email in Gmail.' It clearly distinguishes itself from sibling read/search/label tools by emphasizing the draft creation and explicitly stating 'Does NOT send.' This leaves no ambiguity about the tool's primary function.

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 concrete usage direction: 'Pass the original message's from/subject/threadId/message_id plus your composed reply body.' It also explains when it is not appropriate by noting 'Does NOT send,' which helps differentiate from a send operation. It stops short of naming alternatives or explicitly saying 'use X instead,' so it gets a 4 rather than a 5.

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