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

Annotations only declare the safety profile (readOnly=false, destructive=false, openWorld=true), so the description carries the async burden and does it well: durable task vs job ID, the polling endpoint per client type, and the idempotency-key recovery path for uncertain responses. It stops short of stating rate limits, whether the queued turn can be cancelled, or what happens on conflicting concurrent turns.

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?

Five short sentences, front-loaded with the action, then retrieval path, then recovery and exclusion rules. No filler and no repetition of the title.

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 an output schema present the description needn't describe return payloads, and it instead covers the integration path an agent actually needs (poll, then fetch the saved plan). The remaining gap is the two enums (decisions, execution_policy) whose behavior is never explained outside the schema.

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 only 43% across 7 parameters, so the description should compensate. It adds real meaning for idempotency_key (reuse only on timeout; new key for a new operation) and reinforces the concept_image restriction, but says nothing about decisions, execution_policy, message, or study_id, leaving several parameters to the schema alone.

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 a specific verb and resource ("Queue one research-design turn") and immediately frames it as the asynchronous counterpart to customize_study, so an agent can place it against the synchronous sibling. "Research-design turn" is somewhat domain-specific jargon, but the polling follow-up clarifies the intent.

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 explicit branch conditions: 2026 Tasks clients poll tasks/get, other clients poll get_customize_study_job, then call get_study for human review. It also names the exclusion case (native concept images require the synchronous customize_study) and the idempotency_key reuse rule, so the agent knows when this tool is and is not the right choice.

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