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

Annotations already declare destructiveHint=true and openWorldHint=false, so safety is known; the description adds real context beyond that — history is preserved, the response is compact and omits the document, the returned version must be reused for the next batch, and tracking has no expiry. It stops short of saying what happens on partial failure or whether a 300-op batch is all-or-nothing (atomicity is claimed only for replace_group).

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?

Operation semantics are front-loaded after the opening sentence, which is good, but the editor-behavior aside ('un groupe... se sélectionne et se déplace d'un geste... sans changer la géométrie ni le rendu') is product rationale rather than invocation guidance. The description is also entirely in French while the name, schema, and sibling tools are English, which adds cognitive friction for an English-context agent.

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 destructive 300-operation batch mutation with no output schema, the definition covers the operation vocabulary and version-reuse loop well. It leaves out error/partial-failure semantics, how results are reported per operation, and whether history preservation makes the whole call reversible — gaps that matter more given no output schema exists to compensate.

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 0%, so the description carries the burden. It does well on the operation payloads — explicit value shapes for group_blocks ({blockIds: [au moins 2 blocs de la section], groupId?}), ungroup_blocks ({blockIds}), and replace_group ({blockIds, block}) — but says nothing about pageId, and only obliquely implies expectedVersion via 'la version retournée'. Two of three top-level parameters remain semantically under-explained.

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 opening states a specific verb and resource: apply up to 300 grouped corrections to the draft, and it enumerates the exact operation set (update_block, add/remove_block, update_section, replace_group, group_blocks/ungroup_blocks). This is well beyond tautology. It never names the sibling it differs from (e.g., apply_page_changes, update_page, batch_update_pages), so an agent cannot tell from the text alone why it should pick this over those.

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 is implied rather than stated: 'corrections groupées au brouillon' and 'Utiliser directement la version retournée pour le lot suivant' sketch a batch-repair workflow, and 'Ne retirer que les blocs remplacés, garder tous les autres' gives one concrete do/don't for replace_group. There is no explicit when-to-use versus apply_page_changes or how to break work across batches vs a single call.

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