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

Annotations already declare readOnlyHint=true, openWorldHint=true, idempotentHint=true, and destructiveHint=false, so the safety profile is covered. The description adds genuine behavioral context beyond that: the Lucene-style query mechanism, AND/OR clause combination, field-value semantics (e.g., 'etatAdministratifEtablissement: A=active, F=closed'), and an enumeration of return fields. No contradiction with annotations.

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?

Dense but every sentence earns its place: purpose, filterable fields, combination operators, return fields, then sibling routing. The opening sentence front-loads the core scope. It is longer than a minimal description, but the length is justified by the query-language complexity and the multiple filter dimensions an agent needs to know.

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?

The definition is complete for a search tool: an output schema exists (so return values are formally specified), annotations cover the read-only/idempotent profile, the schema documents all three parameters, and the description adds query syntax, field semantics, return-field summary, and routing to the by_siret sibling. Minor gaps — no explicit pagination default guidance and no mention of the company-search sibling — keep it just shy of a 5.

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 100% and the q parameter's schema description is itself exhaustive — it already lists every filter example, the A/F status codes, and the AND/OR combination syntax. The tool description largely mirrors this, adding only marginal glosses like '47.11B for supermarkets' and the concept that these are physical sites. With the schema doing the heavy lifting, the description adds meaningful but limited incremental value, keeping this at the baseline 3.

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, resource, and scope: 'Search French establishments (physical sites, branches, stores) in the INSEE Sirene registry using a Lucene-style filter query.' The parenthetical 'physical sites, branches, stores' actively distinguishes establishments from the sibling company-level tools (insee.company.search), so an agent can tell them apart. The query-mechanism detail ('Lucene-style filter query') further pins down what this specific tool does.

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?

Explicitly names the alternative with the selecting condition: 'Use insee.establishment.by_siret to retrieve full details for a specific SIRET.' This tells the agent when to prefer the sibling. It doesn't explicitly address distinguishing from insee.company.search, but the opening's establishment-vs-company framing covers most of that ambiguity implicitly.

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