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

Annotations already cover the safety profile (readOnlyHint/destructiveHint false, closed-world), and the description reinforces it with "Lecture seule". It adds genuinely new context: the numeric thresholds applied, and the important caveat that results are indicative and superseded by a signed Th-U calculation. It omits what happens with missing/incomplete scene data and what the per-criterion output looks like.

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 with the call condition, then criteria, then expected input shape, then the scope caveat — a sensible order with no filler. It is dense (regulatory codes, formula, nested field list in one breath) but each element earns its place.

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?

For a single-object, zero-required-parameter, read-only check with no output schema, the description supplies the criteria, the input shape, and the reliability caveat. The only material gap is that an agent is not told the shape of the returned verdict per criterion or how partial data is treated.

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 nested `doublages` fields are already documented there, so baseline is 3. The description goes further by tying each field to its acceptance criterion (epaisseur_isolant_mm vs >=100 mm, lambda vs R=e/lambda >=2,5), which adds real interpretation value beyond the schema's type information.

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 opens with a specific verb ("VERIFIER") and resource ("LE DOUBLAGE ITI"), then enumerates the exact four checks it performs with code references (DO-EPAISSEUR, DO-RESISTANCE, DO-PARE-VAPEUR, DO-CONTINUITE). An agent can immediately distinguish this from generic sibling checks such as cao_bim_ponts_thermiques or cao_bim_condensation.

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..." gives an explicit trigger condition, and "Indicatif — calcul thermique signe (Th-U) prime" marks the boundary where a certified thermal calculation must be used instead. It stops short of naming sibling tools to prefer for related checks, so no full when-not/alternative routing.

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