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

Annotations already declare the read-only, non-destructive, closed-world profile, so the description doesn't need to establish safety. It adds genuine context: a connected identity may query another account's task, retries should be finite, and missing session/run references, progress, and completion times are not inferred. Some of the text (identifier redaction rules) is reply-formatting policy rather than tool behavior, which weakens focus without causing contradiction.

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 purpose is front-loaded, but the text runs long and mixes tool semantics with response-formatting policy that belongs in a system prompt. Several sentences restate the same redaction rule three different ways. An agent must read past multiple clauses to reach parameter-related content.

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 return values legitimately need no explanation, and the description does cover error handling, retries, identity scope, and media_type inheritance. What's missing is whether model must match the model used at creation and whether the call returns one task or a collection. The presence of extensive formatting guidance adds length but not invocation clarity.

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 67%, with task_id and media_type documented in the schema itself. The description reinforces task_id's scope ('public APiXO ID', cross-account query) and media_type inheritance ('automatically carry speech/voice media_type from creation', 'never infer it from task IDs or URLs'), which is useful. The model parameter remains essentially undefined beyond the pattern and becomes the main gap.

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?

It opens with a specific verb-and-resource pair — 'Get task status and results using the supplied model and task_id' — which is unambiguous. It never names a sibling tool (get_model_schema, generate_task) or states which one retrieves what, so the agent has to infer the split. The remaining ~85% of the text addresses other concerns and dilutes separation from siblings.

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?

'Read-only queries may retry finitely' and 'missing tasks, unavailable models, or query errors must not trigger model switching or generation resubmission' are implicit guidance about which tools NOT to call after a failed lookup. There is no explicit statement of when to choose get_generation over generate_task/get_model_schema. Usage context is present but must be assembled from scattered negative constraints.

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