resend_failed_delivery

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

Annotations already declare this is a mutation (readOnlyHint=false), open-world and non-destructive; the description correctly aligns with that and adds the useful behavioral fact that resend targets quarantined as well as failed items. It says nothing further about side effects an agent would care about for a mutation: whether recipients actually receive a new message, whether the original failure reason is preserved, idempotency of repeated retries, rate limits, or what happens to the failed-delivery record after a successful retry.

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?

A single short sentence with no filler and the essential action front-loaded. It is efficiently written, though the terseness is part of why behavioral and usage detail is missing.

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 two-parameter mutation with no output schema and only partial annotation coverage, the definition is minimally adequate: an agent knows what to call and roughly with what. It does not cover the post-call outcome (new delivery id, updated status of the failed delivery) or retry constraints, which an agent needs to decide whether and how often to invoke it.

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 100%, and the schema already documents both parameters including the max-10 recipient cap and the default to the alias's recipients, so the baseline is 3. The description adds no extra meaning beyond that, such as whether failed_delivery_id must match a particular state or what happens if recipient_ids conflicts with the original recipients.

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 states a specific verb and object ("Retry delivering") plus the qualifying scope "failed or quarantined email," which cleanly separates it from the read-oriented siblings get_failed_delivery, download_failed_delivery and list_failed_deliveries, and from resend_recipient_verification. It stops short of naming those siblings explicitly, so it is clear but not maximally differentiating.

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?

Scope words "failed or quarantined" imply when the tool applies, and the naming contrast with the get/download/list siblings reinforces that this is the acting tool rather than an inspection tool. However there is no explicit statement of prerequisites (does the delivery need to be recently failed? can a non-quarantined failure be resent?) and no named alternative for cases that don't qualify.

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