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

Annotations already declare readOnlyHint=false, idempotentHint=false and destructiveHint=false, and the description is consistent with that (one-shot creation, no update semantics). It adds real behavioral context the annotations can't express: the once-per-conversation rule, the implicit primary 'Title' column, root-folder default, and that isUnique drives dedupe in item.create. It stops short of failure modes (name collisions, max 40 attributes) but is well above baseline given annotation coverage.

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 purpose and intent, and virtually every sentence carries a non-obvious rule (currency mapping, Title column, stage defaults, dedupe). It is a dense single block of prose for a genuinely complex tool, so length is justified, but bullet-style separation of the attributes/options rules would make it easier to scan.

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 7-parameter creation tool with nested attributes, the description covers placement, schema construction, dataType normalization, option/stage handling, uniqueness, return value ('Returns the new sheetId'), and lifecycle (call once). An output schema exists, yet naming the returned sheetId still helps the agent chain into follow-up calls.

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 43%, and the description compensates heavily: it explains what each `intent` enum value means, mandates passing the full proposed schema in `attributes`, tells the agent to use canonical dataType names with a concrete currency mapping example, warns against adding a separate name field because of the literal 'Title' primary column, and gives the `options.items` shape with `stage` semantics. That is meaning well beyond what the schema encodes.

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 ('Create a new sheet with optional initial attributes') and then distinguishes the two modes via `intent: 'form'` vs `intent: 'sheet'`. It also names the sibling tools used for follow-up changes (sheet.add_attribute / sheet.update_attribute / sheet.remove_attribute), so an agent can separate creation from mutation without opening other schemas.

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?

Explicitly says 'Call this exactly once per conversation' and routes subsequent edits to the named sibling tools, which is exactly the when/when-not guidance needed. It also covers placement prerequisites (use workspace.structure for spaceId/folderId), the no-folder default, and the panel-invoked exception.

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