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

With no annotations, the description carries the full behavioral burden and does so well: it enumerates the state machine (running | done | failed | killed | timeout), defines the done flag, discloses the 8 MiB retention cap with older output dropped and reported in *_dropped, notes the token requirement, and advises polling cadence.

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?

Densely front-loaded: the return shape leads, followed by the offset protocol, then polling advice, retention, and auth. Every sentence carries actionable information with 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?

There is no output schema, so the description must describe return values and it does (state, exit_code, stdout/stderr, done, dropped output). Combined with the schema's id patterns and the usage protocol, an agent has everything needed to call and interpret this tool.

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 50% (task_id and sandbox_id documented, the two offset params are not), but the description compensates by explaining offset semantics — begin at 0 and advance using the next-offset values from the prior response. Minor friction: it names output fields (stdout_next_offset) rather than the input param names (stdout_offset).

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+resource ('Status and NEW output of a background task') and immediately enumerates the returned fields (state, exit_code, stdout/stderr). This cleanly distinguishes it from siblings like kill_task, run_code, and exec_command.

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 concrete usage context: start with offsets 0, then pass stdout_next_offset/stderr_next_offset, and poll every few seconds since each call is quick. It doesn't state explicit exclusions (e.g., when to stop polling or use kill_task instead), but the when/how-to-use guidance is clear.

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