search

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

Annotations already cover the read-only safety profile, and the description adds genuinely non-obvious behavior: results are relevance-sorted, responses are cached client-side for 24h, and limit/offset are applied client-side rather than server-side. The 24h cache is an important staleness fact an agent would not otherwise know. It does not describe the shape of a hit, but that is partly a return-value concern.

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?

Front-loaded with purpose, then a compact bulleted parameter list and a worked example — easy to scan. The examples in the bullets duplicate the schema's own examples, so a sentence of the content is not strictly earning its place.

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?

There is no output schema, so the description carries the burden of explaining what a result looks like, yet it never says what fields a hit exposes or how hits are typed. Caching, sorting, and client-side pagination/truncation are well covered, but the return-side gap keeps it at minimum viable.

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 description coverage is 100%, so every parameter is already documented, and the description's parameter notes largely restate the schema (query examples, type filter, paging math). It adds the clarification that type matching is case-insensitive and that paging is client-side, but says nothing about 'raw', which is the largest-token-impact parameter.

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?

Opens with a specific verb+resource pair ('全局搜索 CDISC Library') and enumerates what is searchable (标准、数据集、变量、术语). It is clearly distinguishable from the get_* siblings, which are direct-lookup tools, though that contrast is only implicit rather than stated.

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 conveyed only through examples ('query=AVAL, type=Analysis Variable, limit=10'); there is no explicit statement of when to prefer search over get_variable/get_dataset or when the tool should not be used. An agent can infer the intent from context but the guidance is not spelled out.

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