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

Annotations already declare readOnlyHint and idempotentHint, so the safety bar is lower, but the description adds real behavioral context beyond those flags: empty arrays mean no activity, not error; the key list is not closed and must be read from the response; and the query reflects live organizational-wide state across all inboxes in a shared account.

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 longer than average, but almost every sentence adds distinct value: key examples, the empty-array semantics, the unpinned shape warning, and the scope caveat. It loses one point for slight redundancy between 'organization-wide view spanning every inbox' and the following 'reaches inboxes other callers created' sentence, which could be tightened.

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 metrics tool with a rich output schema and high schema coverage, the description fully covers what an agent needs to call it correctly: what it returns, the meaning of empty data, the non-fixed key set, the shared-account scope, and the explicit routing to the sibling when inbox-level metrics are needed.

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%, so the schema already documents all seven parameters including start/end/limit/period/descending/event_types. The description adds useful context about the return shape and metric key names but does not need to explain parameters any further; the baseline of 3 is appropriate.

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 and resource: 'Returns aggregate counters for the whole account as a flat map of counter name to an array of data points.' It clearly differentiates itself from the inbox-scoped sibling by naming the twin explicitly and drawing the account-vs-inbox distinction.

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 explicitly states when to use this tool versus the alternative: 'the inbox-scoped twin `get_agentmail_inbox_metrics` is the one to use when a single inbox is meant.' It also warns that the shared account means this reaches inboxes other callers created, giving concrete guidance about side effects of scope.

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