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

Annotations already cover the safety profile (readOnlyHint true, destructive false, openWorld false), so 'Lecture seule' earns little, but the definition adds real behavioral context: 'Indicatif — charte de modélisation BIM signée prime' tells the agent the verdict is advisory and can be overridden by a contractual document. No information on result shape or behavior when a parameter is absent.

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?

A single front-loaded trigger sentence followed by scope, parameter mapping and a caveat — no filler. The all-caps emphasis and repetition of the parameter list in two forms ('PA-…' codes then the sub-key list) is slightly redundant but compact.

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?

Covers what is checked, the read-only nature and the 'indicative' caveat, which is enough to call it. With no output schema and a nested object parameter, it still leaves the agent guessing what a check returns (per-object verdicts? counts? which object in the scene is referenced).

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% at the top level, but the nested properties of 'scene' have no individual descriptions; the definition supplies them by mapping each sub-key (classification, matériau, référence, marquage) to its PA- code, which is genuine added meaning. It still does not say whether 'scene' takes an identifier, a file, or an inline object.

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?

Specific verb-plus-resource (vérifier les paramètres d'objet / qualité LOI) with the four checked families enumerated and mapped to their PA- parameter codes, which makes it identifiable among the many cao_bim_* siblings. It does not, however, name any sibling it differs from (e.g. cao_bim_lod, cao_verifier).

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?

The opener 'Appelez quand vous devez VÉRIFIER…' gives a trigger, but it essentially restates the purpose and offers no when-not conditions or alternative tools for overlapping checks (LOD, nomenclature, intégrité). Usage is implied rather than delimited in a catalogue of ~200 siblings.

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