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

Annotations already declare readOnlyHint=true and idempotentHint=true; the description adds the crucial approximation nature, the output shape (point central, fourchette, niveau de confiance), and that it substitutes for a missing RIS rather than the reverse. No contradiction with annotations. A small gap: it doesn't state whether results vary across calls or detail the confidence computation, but the safety profile is covered by annotations.

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 core purpose is front-loaded before the caveat and routing. It is somewhat long, but every clause earns its place: purpose, the critical 'not certain' caveat, sibling routing, and source citations. No filler or repetition of schema content.

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 7-parameter, 3-required estimation tool with no output schema, the description covers the method, the return value shape, the substitution caveat, and the sibling alternatives. It explains why a fourchette/confidence level is returned rather than an exact value. Minor omissions (precise confidence definition, traitement_indiciaire handling for fonctionnaires) are partially covered by schema, so the package is adequate.

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 description coverage is 100%, so the schema already documents all 7 parameters, their defaults, minimums, and the accepted statut values. The description adds some semantic color (SAM plafonné au PASS, décote, and how trimestres_acquis/annee_debut_carriere tighten the range) but that mostly restates what the schema param descriptions already convey. Baseline 3 is appropriate since the schema carries the load.

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 ('estimation'), resource (pension), and the exact inputs (statut, revenu, âge, âge de départ) with the method (SAM plafonné au PASS, décote). Explicitly distinguishes itself from the exact-calc siblings by declaring it is 'PAS un calcul certain' and naming the alternatives, so an agent can tell it apart without opening schemas.

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 explicit when-to-use (relevé de carrière RIS manquant) and when-not-to (relevé connu → prefer exact), then maps each condition to a named sibling: régime en points → retraite_pension_regime, fonction publique/IEG → retraite_pension_annuites, multi-régimes → retraite_pension_totale. Nothing is left to inference.

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