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

Annotations only say readOnly=false, destructive=false, openWorld=true; the description goes well beyond that by declaring idempotency, that the tag is back-applied to historical rows, that contacts are linked without altering consent, that the workflow is created in draft, and that reprocessing triggers no retroactive actions. Those are exactly the side-effect details an agent needs before invoking a write tool.

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?

Four short, front-loaded sentences each carrying a distinct guarantee (idempotency, historical rows, consent preservation, draft state, no retroactivity). Minor overlap between 'de manière idempotente' and 'ne déclenche aucune action rétroactive' keeps it from being maximally tight.

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 one-parameter but multi-effect mutation tool with no output schema, the description covers the important consequences (tag backfill, contact linking, draft workflow, no retroactive actions). It leaves out failure conditions and whether anything is returned, which an agent would still have to discover by calling it.

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 0% — the single 'id' parameter has only type/length constraints and no description. The description implies the id refers to an existing collection ('d'une collecte existante') but never states that explicitly or clarifies behavior when the id does not resolve, so it only partially compensates for the blind schema.

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 pair ('préparer ou réparer') and a specific resource ('le tag et le workflow d'une collecte existante'), plus the idempotent nature. It is clearly distinct from create_collection or update_collection, but it never names those siblings or explicitly contrasts itself with them.

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?

The phrase 'préparer ou réparer ... d'une collecte existante' implies when to use it (an existing collection lacking/needing repair of tag and workflow), but there is no explicit when-not, no prerequisites (e.g., does the collection need to exist in a certain state?), and no named alternatives such as update_collection or describe_collections.

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