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

Annotations already cover safety and idempotency, so the bar is lower; the description still adds real context: the 24h bulletin scope, inclusion of suivi texts, and the feux_detectes satellite line with an explicit false-positive caveat and the mandatory emergency-number message. It does not state how often the bulletin refreshes or how stale the data can be.

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-loads the authoritative source and payload in the first clause and keeps example queries compact. The closing fire-safety sentence is mandatory content but makes the block dense; nothing is truly redundant.

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 must convey the return shape, and it does: color per zone for each phenomenon plus follow-up texts and an optional feux_detectes line. Only freshness/cadence of the bulletin is left unspecified.

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 single parameter already documents that vide = toute l'île. The description repeats that rule and adds only the resolution behavior (a place maps to its commune's vigilance zone), which is a mild gain over the schema.

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 resource (Météo-France official vigilance colors by zone plus cyclone alert) with the exact phenomena covered (fortes pluies/orages, vents forts, vagues-submersion) and the 24h bulletin window. An agent can distinguish it from meteo_actuelle, prevision_meteo, cyclones and maree without opening any schema.

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 concrete trigger phrasings («on est en vigilance orange ?», «la mer est dangereuse ?») and the avec/sans lieu branching rule, which is clear operational guidance. It stops short of naming when to prefer a sibling (e.g. prevision_meteo for a multi-day outlook, cyclones for the storm track), so the boundary is implied rather than explicit.

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