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

Annotations flag destructiveHint and this description earns credit by specifying exactly what is destroyed ('Delete it and all its data'), the safety guard (refused while a sandbox holds it), and the retention carve-out for backups. It also discloses non-obvious runtime behavior: single read-write mounter at a time, read-only snapshot copies, and off-host daily backups with backedUp flipping only after a checked copy. Loses a point only because the 'mode "snapshot"' and 'volumes' attach mechanics reference surfaces not present in this schema, leaving their applicability unclear.

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-loaded with the resource definition, then action-by-action in the same order as the action enum, so an agent can scan to the relevant clause. The prose is dense and telegraphic ('Give sizeMiB, or fromBackup to restore one') but nearly every sentence carries an operational fact.

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?

With 10 parameters at 30% schema coverage, no output schema, and a destructive annotation, the description does the heavy lifting on lifecycle, attachment, backup, and deletion semantics. Remaining gaps are the undocumented state/region parameters and the dangling references to volumes/mode that this tool's schema cannot satisfy.

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 only 30%, so the description must compensate; it does explain sizeMiB-vs-fromBackup exclusivity, id for get/delete, and cursor round-tripping via nextCursor. However it says nothing about the state filter values (creating/ready/failed/deleting), region, or idempotencyKey, and it introduces 'volumes: [{volumeId, path}]' and a snapshot 'mode' that have no corresponding schema properties, which risks misleading the agent.

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?

Opens with a concrete definition of the resource ('Persistent disks that outlive sandboxes') and then enumerates the four dispatched actions (create/list/get/delete) with their inputs. An agent can tell this manages volume lifecycle rather than sandboxes or snapshots, though it never names the nearby siblings (runtime_volumes_backup, runtime_snapshots) it might be confused with.

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 per-action usage hints (give sizeMiB or fromBackup to restore; pass nextCursor back as cursor; delete is refused while a sandbox holds it), which is more than implied. But it never states when to choose this tool over runtime_volumes_backup or runtime_snapshots, and no preconditions or alternatives are spelled out.

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