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

The annotations only say readOnlyHint=false and destructiveHint=false, which alone would leave the agent guessing whether the write is immediate. The description closes that gap decisively: it does not touch Xero, it queues a pending change, a human must approve via the response's approvalUrl, and the agent cannot self-approve. It also discloses the refusal behavior for conflicting phone/mobile values in NZ/UK, which is behavioral information available nowhere else.

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?

The key operating model (proposal, not execution) is front-loaded in the first two sentences, which is exactly right for a tool whose biggest misconception is that it writes to Xero directly. It is dense and long, with capitalized emphasis and some duplication of the phone/mobile and address rules that already appear in the schema, which costs a point but not more.

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?

With no output schema, the description carries the burden of explaining the return value, and it does by naming the approvalUrl field and the pending-change lifecycle it participates in. Combined with the region matrix and the employeeId provenance rule, an agent has everything needed to call this correctly and to know what happens next.

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 per-field meanings are already documented, including the AU-vs-NZ/UK phone/mobile split. The description still adds value the schema lacks: the 'at least one mutable field' requirement (the schema only requires employeeId), and the note that only supplied fields change while everything else is preserved. It largely restates the region rules already present in the phone/mobile schema descriptions, which keeps it 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?

States a specific verb and resource (proposes a change to ONE employee) and enumerates the exact mutable field set, explicitly bounding it with 'STRUCTURALLY NOTHING ELSE.' An agent can immediately distinguish this from update_xero_contact or the create_* payroll tools 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 explicit prerequisites (employeeId must come from list_xero_payroll_employees, never invented), an invocation constraint (provide at least one of the six fields), and negative guidance on what is out of scope (bank details, tax, pay templates, KiwiSaver, dates) because no such fields exist by design. It also explains the downstream approval step that follows, so the agent knows this is not a terminal call.

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