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

Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false and openWorldHint=false, so the safety profile is covered. The description adds context annotations cannot express: coverage is limited to writes through this API (write, edit, editMany, rename, delete) and excludes SFTP or app-side edits, and an empty list means nothing to undo. That exclusion is the single most decision-relevant fact here; only the return-entry shape is left implicit.

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 sentences, front-loaded with the action and ordering, then triggers, then the coverage caveat. No restatement of the title and no filler; the caveat sentence is long but each clause (what is tracked, what is not, empty-list meaning) carries distinct information.

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 two-parameter read-only list with no output schema, the description covers purpose, triggers, coverage limits and the restore handoff. It stops short of describing what a version entry contains beyond the versionId needed for files.restore and the newest-first ordering, which is the only remaining omission.

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 50%: 'path' is documented inline as volume-root relative with an example, but 'id' has no description in either the schema or the description. The description reinforces that the pair targets a single file, but it never explains what 'id' identifies, which is a real gap for a required parameter.

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 ('List the saved previous versions of ONE file') plus the ordering ('newest first'), and the 'ONE file' emphasis distinguishes it from bulk listing siblings like files_list or backups_list. An agent can identify the tool without opening the 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 three concrete trigger conditions (a change broke something, the user wants to undo, before overwriting a file you did not write) and names the follow-up action, passing a versionId to files.restore. This is explicit when-to-use plus the correct next tool, leaving little to inference.

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