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

Annotations only mark it as non-read-only; the description goes further and explains the side effects (refreshes worker status, may persist completion or issue a failure refund), which justifies readOnlyHint=false. It still does not say what is returned while the task is still pending or what happens if the task is not owned.

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 with zero filler: purpose first, side effects second, operational constraint last. Every clause carries actionable information.

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 and a mutating tool, the description covers return content, side effects, and polling cadence, which is most of what an agent needs. It omits what a non-terminal task response looks like and how ownership failures surface.

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 0%, so the description must compensate for the single task_id parameter. 'An owned task' does imply the identifier must reference a task the caller owns, but no format, sourcing, or invalid-id behavior is described (the uuid format is carried by the schema).

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 ('Check an owned task') plus the payoff ('return completed audio links'), which distinguishes it from list_conversions and the preparation tools. It never names a sibling explicitly, so the boundary with list_conversions is inferred rather than stated.

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 operational guidance: poll roughly every 10 seconds and do not submit a new paid task while waiting. That tells the agent how to use it in a polling loop, but there is no explicit statement of when this tool is the wrong choice versus list_conversions or get_profile.

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