search

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

Annotations already cover the safety profile (readOnly, non-destructive, closed-world), so the bar is lower, and the description still adds substantive behavior: 'No answer generation', the exact return shape (source spans, PDF pages, section headings, result IDs), and the 10-item cap on paper_ids. The unexplained idempotentHint=false is the main omission.

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?

Four short sentences, front-loaded with the core action and free of filler; the 'no answer generation' disclaimer sits last where it functions as a caveat. Slightly telegraphic phrasing ('source-first') costs a bit of immediate clarity.

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 only two parameters, no output schema, and annotations already declaring the safety profile, the description closes the important gaps: it describes the return payload, forbids answer generation, and constrains both inputs. Nothing an agent needs to invoke it correctly is missing.

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 0%, so the description must compensate, and it does for both parameters: query must use faithful English scientific terminology without presupposing answers, and paper_ids is optional, scopes the search, and is capped at 10. It does not say where paper_ids values come from (presumably list_papers), which keeps it from 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?

The description gives a specific verb+resource ('Retrieve source-first scientific evidence') and immediately clarifies that it returns spans, pages, headings, and IDs rather than answers, which separates it from fetch and audit_answer. The name 'search' alone is generic, so the description is doing real work, though it never explicitly names the sibling it differs from.

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?

Usage is implied rather than stated: 'Result IDs for fetch' hints at a search-then-fetch chain, and the guidance on phrasing Chinese questions in English terms is operational advice. But there is no explicit when-to-use versus list_papers or citation_network, and no stated preconditions or exclusions.

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