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

Annotations only declare readOnlyHint=true and openWorldHint=false; the description adds far richer behavior: the three-status result model, the nearMiss flag with a concrete example ('JJ Dungun' for 'PERNIAGAAN JJ'), what data comes back in the resolved case, and the hard constraint to ask rather than guess. Nothing contradicts the read-only annotation — resolving is consistent with not mutating state.

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?

Three sentences: purpose, status model, and the nearMiss/ask-user rule. Each adds distinct value and the purpose is front-loaded, but the closing 'Never guess among candidates — ask the user' mildly restates the earlier ask-the-user instruction, so one clause is slightly redundant.

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 minimal schema (1 param, no output schema) with nuanced behavior, the description is near-complete: it covers every outcome the agent will face, what data comes back in the resolved case (id + A/R balance), how to recognize shorthand matches, and what to do in each branch. The only minor omission is an explicit next-step for the 'none' status, but the confirm-before-creating rule effectively covers it.

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 0% with only a bare 'name' string field, so the description must compensate. It does: the first sentence establishes that the parameter is a customer name as written on an invoice draft, and the nearMiss discussion clarifies fuzzy and prefix-matching semantics. Only a formal 'name: ...' parameter breakdown is missing, which matters little for a single obvious parameter.

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 opens with a specific verb and resource: 'Resolve a customer name for an invoice draft against this company's real customers.' It enumerates three distinct outcomes (resolved, ambiguous, none) with their payloads, and the customer-specific scope clearly distinguishes it from sibling tools resolve_vendor and resolve_product.

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?

The description gives strong decision rules: report nearMiss matches as likely shorthand and ask the user, only file as new on explicit confirmation, and never guess among candidates. It does not explicitly name alternatives (e.g., use resolve_vendor for vendors), but the customer-scoped framing and the detailed handling rules make when-to-use clear. An explicit when-not-to-use note would fully close the gap.

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