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

Annotations already declare readOnlyHint/destructiveHint/openWorldHint, so the safety profile is covered; the description adds what each mode actually computes (quantitatif x taux unitaires, cost cumulated over time, cash-flow synthesis) and the default branch. The trailing 'Lecture seule' merely restates the readOnlyHint, but the mode-level behavior info is genuinely additive.

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 dense sentence that front-loads the trigger condition and then branches through the three actions with their parameters. Nothing is wasted, though the colon-chained structure is heavy and the redundant 'Lecture seule' tail could be dropped.

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?

With no output schema, the description usefully characterizes what each action returns; safety is covered by annotations and all four parameters are schema-documented. Remaining gap is the nested 'scene' object, whose expected structure is only hinted at ('murs, boites, ouvertures...') in neither description nor schema detail.

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 schema already documents action, phase, and debut, including defaults. The description adds only marginal semantics (that phase filters tasks by BEP stage), so the baseline 3 holds; the 'scene' nested object's expected contents remain equally thin in both places.

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 states a specific verb+resource pair ('estimer le cout (5D) et le planning de construction (4D) d'une scene CAO') and enumerates the three distinct output modes. It does not name or rule out the closely-named siblings (cao_4d, cao_bim_4d_cpm, cao_bim_planning_cpm), so the agent still has to infer the boundary.

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 estimer...' gives an explicit invocation trigger, and each action branch is tied to the situation it serves (bilan = filtered tasks, courbe_s = cumulative S-curve, tresorerie = cash flow). Missing is any statement of when NOT to use it or which sibling tool handles adjacent costing/planning requests.

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