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

The annotations already declare this a safe, idempotent, closed-world read, so the description does not need to restate that. Instead it adds behavior the annotations cannot express: results are newest-first, each row carries a latest-message preview, and paging works by feeding nextBefore back into `before`. The note that 'conversation text in the result is written by website visitors' is a genuinely useful trust signal for an agent reading untrusted content.

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?

Four tight sentences, front-loaded with what the tool returns before moving to filters, paging, and the data-provenance caveat. Slight redundancy with the schema on the before/nextBefore mechanics, but 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?

An output schema exists, so return values need no explanation, and the annotations cover the safety profile. The description fills in ordering, preview content, paging, and the untrusted-text caveat. The only real omission is the result-count default and ceiling, which the agent would need the schema to discover.

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 75%, so the schema already documents before, status (with per-value meanings), and agentId. The description restates the before/nextBefore paging loop and the handoff_requested meaning rather than adding to them, and it never mentions limit, the one undocumented parameter with a default of 10 and a cap of 25. That leaves it at the baseline rather than compensating for the gap.

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 and resource ('List recent conversations between visitors and the account's agents') plus the ordering ('newest first') and the shape of each item ('a preview of the latest message'). That is enough to separate it from get_conversation, which retrieves a single conversation, without opening either schema.

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 a concrete use case rather than a generic one: 'status handoff_requested finds visitors waiting for a person' tells the agent exactly which filter answers a common question. It also says filtering is possible by agent or status. It stops short of naming an alternative tool or stating when not to use this one, so it does not reach a 5.

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