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

Annotations already declare destructiveHint=true and idempotentHint=true, so the description earns credit for going further: it says every copy of the message text is deleted, that opt-out records and billing are retained, and that the action cannot be undone. The retained-data detail is exactly the kind of context annotations cannot carry.

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?

Three short sentences, no filler, with the destructive scope front-loaded and the retention caveat immediately after. Every clause carries information an agent needs.

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 single-param destructive tool with annotations and no output schema, this covers eligibility, blast radius, retention and irreversibility. It stops short of saying what happens to in-flight or scheduled messages, which matters given cancel_scheduled_message exists as a sibling.

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?

Single required parameter with 100% schema description coverage ('Message id (msg_...)'), so the schema already carries the load. The description adds no additional meaning about the id or which message states are valid inputs beyond the finished-state scope already noted.

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 (Delete) plus the resource and, crucially, the eligible states: 'finished message (delivered, failed or received)'. That scope qualifier distinguishes it from queued/scheduled deletion, though it never names a sibling like redirect_to redact_message or cancel_scheduled_message.

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?

'for a privacy request' gives the intent context for using it, so the when-to-use is implied rather than stated. There is no explicit exclusion ('not for scheduled messages') and no pointer to redact_message or forget_identity, which an agent facing a privacy request would plausibly weigh.

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