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

Annotations already declare readOnlyHint/idempotentHint/non-destructive, so the safety burden is covered; the description adds meaningful context beyond that by flagging that the euro figures are estimates (passages × panier moyen) and that reward cost is left unquantified when cout_recompense is omitted. It stops short of describing response shape or whether relaunch results depend on prior configuration, but the estimate caveat is genuinely useful.

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 metrics list is long but every item is a distinct output the agent may care about, and the usage questions plus the tableau_de_bord exclusion are packed into the final two sentences. It is dense rather than padded, though the opening sentence runs on before the reader learns what is actually returned.

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?

With no output schema, the description carries the return-value burden and does so by naming each computed block, including the estimation method and the un-priced-reward edge case. For a zero-required-parameter analytics tool with six optional inputs, nothing needed to call it correctly is missing.

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%, so the baseline is 3; the description lifts it by explaining the purpose of the two override parameters — panier_moyen and cout_recompense exist so the calculation can be re-run with different assumptions, and omitting cost means rewards aren't priced. That is business meaning the schema's per-field text only partially conveys.

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 and resource — "Dit si la carte de fidélité rapporte" — then enumerates the exact metrics returned (retention share, return visits, euro value, median interval, rewards cost, pending rewards, relaunch outcome). It explicitly distinguishes itself from tableau_de_bord, which supplies raw figures instead of the profitability reading.

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?

Names the concrete questions that select this tool ("est-ce que ça vaut le coup ?", "ça me rapporte combien ?", "mes clients reviennent-ils ?") and routes the agent elsewhere for plain period stats via "Pour les chiffres bruts d’une période, tableau_de_bord." The when-to-use and the alternative are both explicit.

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