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

Annotations already cover readOnly/destructive/openWorld, so the bar is lower; the description still adds the important caveat 'Indicatif — etude solaire thermique signee prime', telling the agent the verdict is not authoritative for subsidy eligibility. It does not describe the result structure (pass/fail per criterion), which keeps it below 5.

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-loaded trigger followed by the criteria list, then the data shape, then the caveat — a logical order with no filler. It is dense (field names are restated from the schema), but each clause carries the rule that will be applied.

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?

There is no output schema, and the description never says what a call returns (per-criterion verdicts? an overall pass/fail? a report?), which matters for a check tool. The criteria list partially compensates, and the 'indicatif' caveat is helpful, but the return behaviour remains undocumented.

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% (baseline 3), and the description goes beyond it by pairing each nested field with the rule it is judged against (surface_capteurs_m2 >= 2 m2, orientation not north, appoint declared, taux_couverture_pct >= 50 %), including units. That is real semantic value over a bare nested-object 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?

The description uses a specific verb ('VERIFIER LE CHAUFFE-EAU SOLAIRE INDIVIDUEL') on a specific resource and enumerates the exact checkpoints (CE-SURFACE, CE-ORIENTATION, CE-APPOINT, CE-COUVERTURE) with thresholds. It is clear what the tool does, though it never names near-neighbours like cao_bim_production_ecs or cao_bim_energie, so differentiation must be inferred from the scope.

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?

'Appelez quand vous devez VERIFIER...' is an explicit call condition, and the enumerated criteria make the use case concrete. However, it gives no when-not guidance and no alternative tool to prefer for thermal-solar studies proper (only a caveat that a signed study is needed for the prime).

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