Deep Research（重型）

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

Annotations only declare readOnlyHint and openWorldHint, so the description's disclosure that this is a slow, heavyweight call with cost/rate implications (避免频繁调用) is genuinely additive. It omits anything about result shape or session behavior beyond what the backend_uuid schema description already carries.

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?

Two compact sentences with the heavy-call warning and the preference for lighter siblings front-loaded; there is no filler and every clause carries an instruction.

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, the description should say something about what a research call returns, but it does not, and strip_thinking remains undocumented. The cost/latency framing and sibling routing are present, so it is adequate but incomplete for a 3-parameter 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 only 33%: backend_uuid is documented in the schema, while query and strip_thinking are not. The description partially compensates by advising how to shape the query (merge sub-questions into one call), but says nothing about strip_thinking, leaving one parameter unexplained in both places.

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 the resource and character of the operation (深度研究, 重型调用) and explicitly positions it against its siblings by naming ask/search. It is clear enough that an agent can distinguish it from perplexity_ask and perplexity_search without opening a schema, though the core verb is partly restated from the tool name itself.

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 an explicit use-only-when-necessary rule, names the preferred alternatives (优先 ask/search), warns against frequent invocation, and advises consolidating sub-questions into a single call. This is a complete when/when-not/alternative routing instruction.

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