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

Annotations already declare readOnly, idempotent, non-destructive behavior, and the description is consistent with that while adding real context: no persistence, native validation, exact before/after preview, and omissions preserved in the patch. It stops short of stating what the preview output looks like or whether the revision token can go stale between read and prepare.

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 compact and has no filler, but the fragment-heavy, telegraphic style pushes comprehension onto the reader and leads with the trigger ('After explicit request') rather than the purpose. Phrases like 'Native validation' are not self-explanatory, and the safety statements are split across the middle and end rather than front-loaded.

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, nested-patch tool with annotations but no output schema, the description covers the key behaviors an agent needs: revision-token sourcing, partial patch semantics, validation, preview, non-persistence, and the human-execution handoff. The remaining gap is the absence of any hint about how to source the revision (which get_ tool) and what the preview itself contains.

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 description coverage is 0%, so the description must carry the parameter burden. It meaningfully explains both: expectedRevision must be copied exactly from the settings read, and patch fields are a partial-update set where omissions are preserved. It does not explain the business-vs-brand `setting` discriminator or any of the individual brand fields (colors, fonts, tone of voice), which remain undocumented in both description and schema.

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 description conveys the operation concretely: read matching settings, copy the exact expectedRevision, supply structured patch fields, run native validation, and produce an exact before/after preview with no persistence. That distinguishes it from the confirm_/get_ siblings in behavior, though the specific resource ('brand guidelines') lives only in the name/title and the telegraphic phrasing makes the central 'prepare a preview' purpose slightly oblique.

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 gives a trigger condition ('after explicit request'), an exclusion ('no publication, send or reissue'), and names the alternative path ('human execution uses show_business_settings_approval with identical input'). What it does not spell out is the relationship to the read tools used to obtain the current revision or to the confirm_* tools that commit the change.

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