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

Beyond the annotations, it discloses a lot that matters operationally: all-or-nothing application ("Every op applies, or none"), optimistic concurrency with named conflicting editors, stable address semantics across moves/renames/chart numbering, and a described response shape ("each cell old → new, results that changed, and the new version"). The one tension is that it advertises `deleteSheet`, `deleteRows`, `deleteCols` and `deleteTable` while `destructiveHint` is false, but the versioned/undoable model partly justifies that annotation, so it reads as a nuance rather than a flat contradiction.

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?

It is appropriately short and front-loads the purpose, but the prose is compressed into clause-heavy parentheticals ("the version (v N) your last get, read or changes showed") that take real effort to parse. The information earns its place; the sentence construction makes it harder to absorb than necessary.

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 an 8-parameter nested-op mutation tool with no output schema, the description covers what an agent needs: versioning contract, atomicity, conflict handling, dry-run semantics, undo, and the shape of the answer. Nothing essential about calling it correctly is missing.

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 100%, so the baseline is 3, but the description genuinely adds meaning the schema does not: `base` is what makes addresses, headers, chart numbers and sheet names resolve consistently, and `undo` is constrained to "goes alone: no base, ops or expect." `note` and `key` are left mostly to their schema text, keeping this below a 5.

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 clause "Changes a sheet" gives a clear verb plus resource, and the rest of the sentence sharpens scope by pinning the edit to a specific read version ("read at 'base': the version (v N) your last get, read or changes showed"). An agent can tell this is the mutation counterpart to `read`/`get`/`changes`, though the description never explicitly contrasts itself with adjacent editing siblings such as `edit_doc` or `apply`.

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?

It establishes the prerequisite (work from a version returned by get/read/changes) and gives a concrete recovery path when the assumption breaks: "a cell someone changed since is refused, naming who, so read again and apply on the new version." It also routes preview and revert behavior to `dryRun` and `undo`. It stops short of explicit when-not-to-use guidance or naming an alternative tool for a different editing surface.

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