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

The annotations already declare readOnlyHint, idempotentHint, destructiveHint=false, and openWorldHint, so the safety profile is already covered. The description adds genuinely useful behavior: polling for output, the cursor lifetime, and the fact that the response reports liveness and exit code. It does not address what happens when the cursor expires or when the sandbox has ended, beyond implying invalidity.

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, each carrying distinct information: what is returned, how the cursor behaves, and the lifetime limit. Nothing is redundant, and the return contract is stated first.

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 the return fields do not need to be explained, and the description covers the call pattern and validity window well. The one gap is that it does not explain wait_ms, meaning that long-polling is left to the schema. For a four-parameter, read-only polling tool, this is nearly complete.

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?

The cursor parameter has no description in the schema, and the description compensates fully by explaining how to start from the beginning and how to advance the cursor. The process_id and sandbox_id semantics are already covered by the schema text. The wait_ms parameter, a long-poll timeout defaulting to 0, is left entirely to the schema, so the description does not mention the blocking behavior.

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 states precisely what the tool returns: the output of a background process after a given cursor, whether the process is still running, and its exit code. It clearly differs from the sibling tools that write or start things, but it never names the alternative, such as rent_start_process or rent_exec. It is more a description of the return payload than of the action, but it is specific enough for an agent to choose this tool correctly.

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 concrete instructions on how to use the tool: omit the cursor to read from the start, and pass the cursor returned by the answer back unchanged to read only newer output. It also gives a lifetime constraint: cursors and process ids are valid while the sandbox runs. It does not say when not to use it, or which tool to use for the same job on non-background commands.

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