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

Adds real behavioral context beyond the annotations: a 6-digit code is texted, the exact consent wording is included, and mandated requests can bypass the code step. However, destructiveHint=true and idempotentHint=false are never explained — nothing says that retrying issues another SMS/code or what state is overwritten. Since annotations cover the safety flags but the description leaves their practical consequences opaque, a 3 is appropriate.

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?

Compact and front-loaded: the core action and the code are in the first sentence, prerequisites and the auto-accept variant follow. The bilingual duplication doubles length but the French version carries all the information and nothing is 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?

Ten parameters, no output schema, and a destructive external side effect. The description never says what the call returns — a code reference, a pending-request id, a send status — which the agent needs since there is no output schema. Everything about inputs and flow is covered; the return side is a genuine gap.

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, but the description adds provenance and ownership meaning the schema does not: 'Le prénom et le cellulaire sont ceux de la personne, pour elle-même' and that the mandate key comes from confirmer_code after proposer_limites. It still does not explain the heures requirement (at least 2 slots derived from moments) or the lat/lng vs lieu choice.

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 ('Prépare la demande chez un commerce', 'texte à la personne un code à 6 chiffres') and explicitly separates this tool from confirmer_code, which only sends the request. An agent can tell from the description alone that this is the code-issuing step in a two-step flow.

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?

Explicit routing: use this to prepare and text the code, then use confirmer_code with that code to actually send. It also names the precondition path ('la clé de confirmer_code après proposer_limites') and the exception where no code is needed (mandate key + accepter_seul within all signed limits). Both alternatives are named by tool name.

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