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

Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false and openWorldHint=true, so safety is covered; the description goes further by disclosing the mandatory-input failure mode (no consumption data → no figures, plus guidance on how to obtain it) and the output scope including a catalog-limited recommendation with the caveat that other models may fit. It does not describe runtime behavior, input-validation errors, or data-format failure handling beyond the missing-data case.

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-loading is good — purpose, then outputs, then the hard requirement, then use cases, then the disclaimer. But the entire content is stated twice (French then English), so roughly half the text is duplication that adds no information for a single-pass reader, and the parenthetical output list is dense.

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?

An output schema exists, so return values need not be explained, and the description covers everything else an agent needs: prerequisites, accepted input sources (GRDF/Enedis CSV, fioul deliveries, France-only scope), intended use cases, and accuracy limitations. Nothing material is missing for correct invocation.

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 the baseline is 3, and the description adds genuine cross-parameter meaning the schema does not: releve_consommation is 'indispensable pour gaz et électricité' while fioul uses livraisons_fioul (at least 3 deliveries), and code_postal drives local weather lookup. That conditional dependency is the main thing the flat schema cannot express.

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 names a precise verb and resource ('Dimensionnement d'une pompe à chaleur (PAC air/eau)') and enumerates the concrete outputs (bilan thermique, déperditions, puissance à la température de base, consommation électrique estimée, économie annuelle, modèle du catalogue). It is unmistakably about heat-pump sizing. It never names or contrasts the one sibling, estimer_aides, so differentiation is left to inference.

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 explicit when-to-use triggers ('quelle puissance de pompe à chaleur pour ma maison, remplacer une chaudière gaz, fioul ou un chauffage électrique, estimation avant la visite d'un installateur') and a hard prerequisite with its consequence ('Un relevé de consommation est NÉCESSAIRE : sans lui, l'outil ne donne aucun chiffre'). It also bounds the scope ('Estimation préliminaire : ne remplace ni la visite technique ni la note de dimensionnement de l'installateur'), so the agent knows when this tool is the wrong answer.

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