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

Annotations already declare readOnly/idempotent/non-destructive, but the description adds a nuance annotations do not cover: 'Aucune lecture, aucune écriture' — it neither fetches nor mutates, it only renders the values it is handed. That pure-presentation, no-computation behaviour is genuinely additive. It stops short of describing output structure or any rendering limits.

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?

Three sentences with the core action front-loaded in caps ('AFFICHE UN ÉCRAN'), followed by use-case context and then the read/write disclaimer. It is appropriately sized and mostly waste-free, though the light/dark mention is a minor loose end.

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 7-parameter tool with nested objects and no output schema, the description orients the agent well on purpose and behaviour but leaves the nested parameter contracts (tableau, graphique, indicateurs.evolution) to the schema. Adequate but with clear gaps given the complexity.

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 only 14% across 7 params with nested objects, so the description must compensate — and it does name the key blocks (titre, indicateurs, tableau, graphique, conclusion). But it gives no format or constraint detail for the nested structures (row/column limits, graph type enum, evolution object), and 'en clair ou en sombre' alludes to a theme option that has no corresponding schema property, adding mild ambiguity.

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 and resource: it renders ('met en forme') an already-computed analysis into a Ficiel-styled screen, and enumerates the rendered blocks (titre, indicateurs, tableau, graphique, conclusion). The phrase 'analyse déjà calculée' and 'ne calcule rien' clearly separates it from sibling compute/read tools such as ficiel_synthese or ficiel_get_stats.

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?

Gives a concrete context ('analyse sur mesure faite de plusieurs lectures : résultats par message d'ouverture, comparaison de comptes ou de séquences, résumé de compte'), which tells the agent when this tool is the right fit. It stops short of naming an explicit alternative or a when-not condition, so it is a clear context rather than full routing.

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