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

Annotations already declare destructiveHint=true, idempotentHint=false and not read-only, but the description adds substantial context beyond them: the change is applied live to real callers at once, all changes are atomic ('applied together or not at all'), they are versioned and reversible via restore_ai_setup_version, and handbook notes are never auto-removed. No claim contradicts the annotations.

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 purpose and the live-immediately consequence, and most sentences carry actionable rules. It is on the long side, with a fully worked temporary-note example that consumes real space, but the content is mostly load-bearing rather than padding.

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 destructive, non-idempotent mutation with a gate parameter, the description covers everything an agent needs: the confirmation protocol, atomicity, reversibility, field-level semantics, and the follow-up call required to undo a note. An output schema exists, so return values need not be explained.

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?

With 50% schema coverage the description compensates well: it enumerates the editable fields, gives the allowed afterHoursAction values, notes aboutYourBusiness is capped at 500 characters, and explains that sendConfirmationEmail takes true/false and what append/remove do semantically (append to end vs remove one exact piece, neither = full replace). It adds meaning beyond the schema's structural definitions, though the field enum itself is already in the 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?

States a specific verb (changes) and resource (the connected account's AI receptionist answer settings), plus the critical scope fact that changes go live for real callers immediately. It is clearly distinguishable from get_ai_setup (read) and restore_ai_setup_version (undo) among the siblings.

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 explicit prerequisites ('Call get_ai_setup first'), a mandatory workflow ('tell the user exactly which settings will change... get their explicit yes; then pass confirm: true'), and names the alternative modes (append vs remove vs full replace) with the condition that selects each. It also routes temporary changes to the append pattern rather than a separate tool.

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