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

Annotations declare a non-read-only, non-idempotent, destructive-flagged write, and the description adds real context beyond them: partial success is possible, results are per-message with codes, unknown outcomes must be read before retrying, and source folders may be stale. The safety claims ('Preserves content and flags; never permanently deletes') clarify the scope of the destructive flag rather than contradicting it — a move relocates rather than deletes content.

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?

Front-loads the operation and batch limit, then alternatives, then preparation, then return shape, then failure policy — a sensible order for an agent deciding whether and how to call. The staccato sentences are dense but each one carries an actionable constraint, with no restatement of the name or redundant framing.

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?

There is no output schema, so the description supplies the return contract itself: 'moved, unchanged, failed or unknown, with code when applicable' and 'one result per message.' Combined with the retry and partial-success guidance, an agent has everything needed to call the tool and interpret its multi-status response.

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?

With 0% schema description coverage the description must carry parameter meaning, and it does for the non-obvious ones: 'expectedFolderIds from search/read' implies an optimistic-concurrency precondition, and 'messageId' sourcing from search/read is stated. It does not explain 'mailboxId' or 'destinationFolderId' beyond their self-evident names, and the 'same connected mailbox' phrase is the only hint about cross-mailbox restrictions.

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 precise verb and resource ('Move 1 to 200 user-selected messages to one folder in the same connected mailbox in one call') with scope limits and a single-folder constraint. It explicitly distinguishes itself from the sibling 'move_message' by naming it, so an agent can route between the batch and single-message variants without opening either 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 an explicit preference rule ('Prefer this over repeated move_message calls'), a prerequisite sequence ('Collect all IDs before moving. Supply each messageId and expectedFolderIds from search/read'), and failure-handling guidance ('Read unknown outcomes before retrying... Never automatically retry the whole batch'). This covers when to use, how to prepare, and what not to do after a partial failure.

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