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

Annotations already declare the write/safety profile, but the description adds two things they don't: the required write scope and conditional creation semantics ('si elle n'existe pas déjà'), which shape an agent's decision to call it. Remaining gap: it never says what happens when the recipe already exists (error, update, or no-op), which sits in mild tension with idempotentHint=false and leaves repeat-call behavior undefined.

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?

Three short sentences, front-loaded with the action and scope guard, then the exclusion and redirect. No restatement of the title and no filler.

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 non-destructive mutation with full annotation coverage and no output schema, the description covers purpose, authorization, existence handling and the out-of-scope scenario. What it omits is the already-exists outcome and any hint about the step structure an agent must supply, which matters given a required array-of-objects parameter.

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?

Top-level schema coverage is 100% and both properties carry descriptions ('Contenu des relances, écrit mot pour mot par vous', 'true = active immédiatement'), so the schema does the heavy lifting and the baseline of 3 applies. The description adds nothing about the nested step items — notably the meaning of a step's 'id' and how delay_hours composes a sequence is left entirely to the reader.

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 resource ('Crée la recette reengagement_48h ... de votre organisation') and immediately resolves the mismatch between the generic name and the single creatable recipe by excluding comment_to_dm. An agent can separate it from ficiel_set_automation and ficiel_get_automations without opening any schema.

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 the use condition ('si elle n'existe pas déjà', with the existence check apparently handled by the tool), the prerequisite ('Portée « write » requise'), and an explicit exclusion plus alternative path for comment_to_dm (configure in Ficiel then pilot with ficiel_set_automation). When-to-use, when-not-to-use and the fallback are all present.

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