Météo mesurée en ce moment à La Réunion

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

Annotations already declare readOnly, idempotent, non-destructive and open-world, so the safety profile is covered. The description adds genuinely useful behavior beyond that: results are dated measurements tagged with station, distance and altitude, plus a microclimate warning when the requested place differs in altitude or windward/leeward aspect, and a radar-rain fallback for places without a rain gauge (Mafate, cirques, Hauts).

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 measured-variables list is front-loaded and every claim is relevant, but the same content is then restated wholesale in English, roughly doubling length with no new information, and the example-question chain is longer than needed. Effective but padded.

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 it: it names the measured fields, the station metadata (distance, altitude), the rain history block, and the microclimate caveat. An agent can decide to call it and set expectations for the response without further information.

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 earns more by explaining that supplying a lieu switches on the pluviometer mode (hourly rain over 12–24 h, time of last rain, total since midnight) whereas an empty value returns a whole-island sweep. That meaningfully extends the schema's one-line free-text definition.

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+resource (measured current weather) and enumerates the exact quantities returned: temperature, rain over the last 6 minutes and last hour, humidity, wind/gusts, pressure, present weather. It also names the sibling it is not (prevision_meteo for future weather), so an agent can separate it from prevision_meteo and historique_meteo without opening a 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 user-question triggers ('quel temps fait-il à Saint-Denis ?', 'il pleut à Cilaos ?', 'combien est-il tombé cette nuit ?') and an explicit exclusion with the alternative: for later today or the coming days, use prevision_meteo. The dual mode of the lieu parameter (with a place = add pluviometer data; empty = island tour) is also framed as usage guidance.

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