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

Annotations only say readOnlyHint=false, destructiveHint=false, idempotentHint=false, openWorldHint=true — a generic write profile. The description adds the essential behavioral fact that this enqueues a pending change requiring human approval and that nothing is written to Xero until then, plus partial-update semantics ('only the fields you provide change, everything else is left exactly as it is'). That is substantial context beyond the annotations and does not conflict with them.

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 most important fact — this is a proposal, not a write — is front-loaded, and the parameter rules follow in a tight sequence. It is a dense multi-clause passage; a small amount of repetition ('it does NOT change anything in Xero itself, and never will on its own') could be trimmed, but no sentence is filler.

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?

There is no output schema, and the description compensates by naming the return field the agent needs (approvalUrl) and the human-approval workflow that follows. For a five-parameter, non-destructive queued write, an agent has everything needed to call it correctly and to explain the outcome to a user.

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 would be 3, but the description adds a real constraint the schema does not encode: a cross-field rule that at least one of name, emailAddress, or phone must be provided, since only contactId is required in the schema. It also reinforces the contactId provenance rule and the omit-to-leave-unchanged semantics for organisation.

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 an EXISTING Xero contact') and immediately clarifies the critical distinction that it does not mutate Xero directly, unlike a direct update. The word 'Proposes' plus the explicit non-mutation framing separates it cleanly from create_xero_contact and update_xero_invoice 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?

Explicit preconditions: contactId must come from list_xero_contacts and never be invented; at least one of name/emailAddress/phone must be supplied; organisation is only needed for multi-org connections. It also explains the downstream approval path (human approves via the connector's approvalUrl, the agent cannot self-approve), leaving nothing to inference.

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