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

Beyond the annotations (readOnly=false, destructive=true, openWorld=true), it discloses three non-obvious traits: the operation is queued/non-confirming, attachments are not forwarded, and the body is quoted. These are exactly the behavioral facts an agent needs before invoking a mutating send.

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?

Front-loaded with the core action, then the caveats, then the operational follow-up. Every sentence carries distinct, non-redundant information with no filler.

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 complex mutating, open-world tool with no output schema, the description covers the async return model, polling, and idempotency well. It omits guidance on senderIdentityId selection and deliverability/error handling, leaving minor gaps but nothing blocking a correct call.

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 only 25% across 8 parameters. The description only hints at textBody ('optional note'), and does not explain to/cc/bcc arrays, accountId, or senderIdentityId, so it fails to compensate for the large coverage gap on a complex parameter set.

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+resource ('Forward a message') and immediately scopes behavior with quoting and attachment exclusion. An agent can distinguish it from send_email and reply_to_email without opening the schema, and it names get_mail_action as the follow-up sibling.

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 post-call guidance ('Poll get_mail_action with the returned operationId') and an idempotency rule ('Never retry an uncertain send with a new operationId'). It does not explicitly contrast when to forward vs. reply_to_email or send_email, so it stops short of full alternative routing.

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