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

With annotations only marking it as non-read-only/destructive/open-world, the description carries the load: read mutates state (marks messages read), read retention is 7 days vs. inbox retention 1-365 days, quotas (64 KB, 1000 unread), webhook delivery is HMAC-SHA256 signed, and each call costs $0.003 USDC via x402 with named chains. These are exactly the behaviors annotations cannot express.

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 purpose and account-free framing are front-loaded, and the dense sentence about limits, retention, and pricing is information-rich rather than filler. It is long, but nearly every clause conveys a distinct operational fact, so the length is defensible.

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 14-parameter multi-operation tool with no output schema and no per-op annotations, the description covers operations, auth (secret), limits, webhook signing, and pricing. The only real gap is that it never sketches what read/peek actually return (message shape, pagination via 'after'), which an agent must infer from the schema.

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 79%, so the schema documents most fields already, but the description adds cross-parameter semantics (op defaulting rules, secret from create as ibs_..., address as inbox_...) and a concrete sample input showing real parameter values. It adds value beyond the field-level descriptions rather than merely restating them.

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 names a concrete resource (a mailbox for AI agents) and enumerates the exact operations (create, send, read, peek, delete) with their artifacts (inbox_... address plus secret). It is unmistakable against every sibling in the list, none of which deals with inter-agent messaging.

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?

It gives explicit use cases ('hand results from one agent to another, collect job results, or let other agents reach yours') and per-operation guidance (read marks read for job-queue sharing, peek only looks). It does not, however, contrast itself with the nearest conceptual sibling, agent_memory, or state when messaging is the wrong choice.

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