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

Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false and openWorldHint=false, so safety is covered. With no output schema, the description usefully carries the return burden: six stages plus the share of the previous stage, response rate and booking rate, which tells the agent what it will receive. It doesn't cover pagination or latency, but that is not needed for a fixed-shape read.

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 with the key action front-loaded in caps ('AFFICHE UN ÉCRAN'), followed by the stage list and the derived metrics. Nothing is wasted, though the stage chain with parenthetical 'critères écrits pour Kord' is somewhat heavy for one sentence.

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 two-parameter, zero-required, read-only analytics tool with 100% schema coverage and no output schema, the description supplies the missing return-shape information (stages, share-of-previous, response/booking rates) and the channel scope. It is complete enough to invoke correctly, with only the sibling-routing gap remaining.

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%, so the schema documents both parameters (channel enum and period_days). The description adds meaning by clarifying 'tous canaux ou un seul' (channel is optional, all channels by default) and that the period applies to the funnel, going slightly beyond the schema's raw field definitions.

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 (affiche) and resource (l'entonnoir de conversion) and enumerates the six funnel stages, so the agent knows exactly what it produces. It situates itself against the Stats screen ('l'entonnoir de l'écran Statistiques, mêmes chiffres'), giving partial sibling differentiation versus ficiel_get_stats, but does not name any sibling tool directly.

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?

Usage context is implied by the analytics framing (period-based funnel figures), but there is no explicit when-to-use or when-not-to-use guidance relative to siblings like ficiel_get_stats, ficiel_get_stats_by_channel, ficiel_compare_periods or ficiel_synthese. The agent must infer which analytics tool to pick.

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