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

Annotations already declare readOnlyHint=true and destructiveHint=false, so the safety profile is covered; the description's 'Lecture seule' merely restates it. It adds real domain context — the acceptance criteria applied (epaisseur >= portee/30, lambda <= 70, h >= L/15, N ELU/sections <= 8 MPa) and the indicative disclaimer — but says nothing about what the response contains (a calc note? a pass/fail list per element?).

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 purpose is front-loaded in capitals, then criteria, then the scene shape, then a one-line caveat — a logical order with no filler. It is dense (a single very long sentence in the middle), but every clause carries actionable engineering information rather than padding.

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 high-complexity whole-structure pre-design with a nested object parameter and no output schema, the description covers input shape and criteria well. The significant remaining gap is the return value: an agent cannot tell whether it gets a report, a per-element verdict, or modified scene data.

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 single parameter is a nested object, so baseline would be 3. The description goes beyond the schema's one-line summary by spelling out the expected sub-structures and keys (boites {role, l, p, epaisseur}, poteaux {section_mm, hauteur_m}, poutres {b_mm, h_mm, portee_m}, voiles {x1,y1,x2,y2, epaisseur_mm}), which materially helps construct the payload.

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 and scope: PREDIMENSIONNER TOUTE LA STRUCTURE d'une scene CAO d'un coup, and enumerates the elements covered (dalles, poteaux, poutres, voiles, contrainte globale). The 'd'un coup' framing implicitly separates it from the many single-element siblings (cao_bim_structure, cao_bim_descente_charges, cao_bim_planchers), but no sibling is named explicitly.

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 PREDIMENSIONNER TOUTE LA STRUCTURE ... d'un coup' gives an explicit trigger condition, and the closing caveat ('Indicatif — note de calcul EC2 signee prime, la decision reste humaine') tells the agent this is not a final design. No when-not guidance or named alternative for single-element preframing is given.

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