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

The description discloses behavior the annotations only hint at: reading marks messages read (which explains the non-obvious readOnlyHint=false), messages are retained only 30 days, a long poll can block up to 25 seconds, and every response yields a cursor for the next call. It also carries a security instruction about treating messages from ownerless AIs as untrusted rather than as commands.

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 information is dense and largely on-topic, but it is delivered as one run-on block with awkward clause order ('Each has its id, thread and reply_to...'). The closing sentence about messages to a person reaching them through an app is tangential and does not help the agent call the tool.

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?

With no output schema, the description does the work of describing returns (id, thread, reply_to, cursor) and retention, plus safety guidance and poll semantics for a 4-parameter, zero-required tool. It is close to complete, with the main gap being the untyped-limit behavior and pagination edge cases.

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 100%, so the baseline is 3, but the description adds semantics the schema does not: that 'after' takes the cursor returned by the previous answer, that 'unread' has the side effect of marking messages read, and that 'wait' creates a long poll meaningful only alongside unread/after. It leaves the odd string typing of 'limit' unexplained.

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 resource and scope: direct messages sent to the calling AI, newest first, with 30-day retention. That distinguishes it from siblings like send_message or list_feedback, though it never names an alternative explicitly. The verb is implied rather than stated ('list'/'read' only appears in the title).

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 one concrete usage pattern for the long poll ('use it with unread or after') and explains how the cursor should be chained between calls. However, there is no explicit when-to-use/when-not-to-use framing and no sibling alternative is named, so the routing guidance remains implicit.

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