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

Annotations only declare a read-only, open-world, non-destructive operation; the description adds substantial behavioral context beyond that: data provenance (satellite measurement vs forecast model), a 4-day forecast horizon, and the fact that results arrive in two separate blocks with measurement date and distance, and wave height/period/direction. It further discloses two unexpected extra blocks (NOAA coral bleaching risk, cetacean approach rules), which an agent would otherwise not anticipate.

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 core capability and the measured/forecast split are front-loaded, and each sentence carries information. It is dense with example questions, which is mostly justified for intent matching, though the parenthetical quoting style makes it longer than strictly needed.

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 does the work of describing returns: two blocks, measurement plus date and distance, wave height/period/direction, and two auxiliary blocks. It omits units (e.g., °C, metres), data freshness, and what happens when a location cannot be resolved, which keeps it from being fully complete for a data-lookup tool.

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?

There is a single optional parameter with 100% schema description coverage, so the schema already explains that lieu accepts a commune, beach, neighbourhood, cirque, summit, coordinates, or empty for the whole island. The description only adds illustrative place names (Saint-Leu, Boucan, Saint-Pierre, Saint-Gilles), which is a marginal increment, so the baseline of 3 applies.

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 the resource and scope precisely (sea temperature and swell at La Réunion, near a beach or coastal spot) and separates two distinct capabilities: satellite-MEASURED water temperature versus model-FORECAST waves over 4 days. Example user phrasings ('la mer est à combien à Saint-Leu ?' vs 'ça surfe à Saint-Pierre demain ?') let an agent distinguish it from generic weather siblings like prevision_meteo or meteo_actuelle without opening the 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?

Usage context is very clear through five example questions that map real user intents onto the tool, and the description distinguishes measured-versus-forecast sub-cases so the agent knows which data it will get. It gives no explicit exclusions or named alternatives (e.g., when to prefer maree for tides or prevision_meteo for air weather), so it falls short of the when-not/alternatives bar.

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