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

Beyond the annotations (readOnlyHint=true, idempotentHint=true), the description discloses important behavioral traits: keyword is a single substring match over title+abstract, matching is accent-insensitive, metadata/filter values use French labels, and results include only a short snippet. This adds context that annotations do not provide and directly informs invocation expectations.

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?

The description is four sentences long, front-loaded with the primary purpose, then provides key usage caveats, and ends with a cross-reference to a sibling tool. Every sentence serves a distinct purpose (scope, search behavior, multilingual tip, result detail guidance), with no redundant phrasing or filler.

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 a tool with 10 parameters and no output schema, the description adequately covers the most critical contextual aspects: the keyword matching behavior, multilingual considerations, French label conventions, and what results contain. It does not explicitly mention pagination defaults (limit/offset) or the exact list of reference types, but those are already present in the input schema, so the description is complete enough for effective use.

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?

The schema already covers 80% of parameter descriptions, but the description adds extra semantics for the keyword parameter (substring match, one term per call, accent-insensitive) and clarifies that reference_type and language use French labels. This goes beyond simply restating schema descriptions, though the high schema coverage means the description does not need to compensate heavily.

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 opens with a specific verb ('Search') and a resource ('academic references') followed by a parenthetical list of reference types (journal articles, book chapters, theses, books, reports), clearly scoping the tool's function. It also distinguishes itself from sibling search tools by specifying the exact search mechanism ('by keyword and metadata') and later clarifying the substring-match limitation, which differentiates it from broader search tools.

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?

The description provides explicit usage guidance: 'search ONE term per call' with a concrete example of a failed combined query ('pèlerinage Mecque'), and advises trying both French and English keywords. It also tells when to use get_reference instead ('use get_reference for the full abstract and bibliographic detail'), which clearly directs the agent to an alternative tool when more detail is needed.

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