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

Annotations only declare idempotentHint=true; the description adds real behavioral context beyond that — that starting incurs credit charges, that it reads files already bound by the prepare step rather than re-uploading, that retries must reuse request_id/task_id, and that no extra confirmation should be requested. It does not discuss latency, async completion semantics, or failure modes, so it stops short of a 5.

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?

Every clause carries operational content and the authorization precondition is front-loaded, but the text is a single semicolon-chained block that mixes preconditions, retry policy, and output routing, which slightly hurts scannability.

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?

For an 8-parameter launch tool with no output schema, the description covers authorization, model defaults, idempotent retry, billing implications, and where progress and artifacts surface, which is close to what an agent needs. It leaves the semantics of task_type, research_area, and the required file set to 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?

With schema coverage at 38% across 8 parameters, the description compensates by insisting the payload be identical to the prepare_research_task call and that model_id may be omitted to accept the server default, and by forbidding file/model changes on retry. task_type, research_area, and file roles carry no added meaning in the description, so not a 5.

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 (启动/执行科研文件任务) and pins its identity by saying it takes exactly the same parameters as prepare_research_task, then explicitly excludes the literature-review use case (不是文献调研替代分析), which routes the agent away from start_research_review. An agent can distinguish it from prepare_research_task and start_research_review without opening any schema.

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 the precondition (任务与积分执行已获授权), the decision rule for optional model selection (未指定模型就采用默认项，不要求选模或重复确认), the retry rule (复用 task_id 与 request_id，不更换文件或模型，不拆成多个扣费任务), and names the alternatives for follow-up (get_research_review / get_research_artifact). Both when-to-use and when-not-to-use are covered.

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