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

With no annotations provided, the description carries the burden. It discloses that metrics are computed on-demand based on available inputs)Skip the format is not described, but the absence of an output schema means the description should at least hint at return structure; it does not, but the partial-input behavior is a good disclosure. Since the tool is non-mutating and purely computational, the lack of side-effect warnings is less critical. However, no annotation context means the description could be more explicit about return format or error handling, hence a 4.

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 description is two sentences: the first lists the metrics and the second explains the partial-input behavior and directs away from self-computation. Every word earns its place, with no fluff. The purpose is front-loaded with the list of metrics, and the usage guidance is concise.

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?

Given the complexity (10 parameters) and lack of output schema and annotations, the description is surprisingly strong. It covers what metrics are returned restricted to input availability, and it explicitly advises against self-computation. The only gap is that it doesn't describe the exact output schema (e.g., JSON structure of metrics and formulas), but since the tool is computational and the description clearly states the provided metrics, an agent can infer the output. For a tool with this complexity, it provides most of the needed context, hence a 4.

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 60%, meaning some parameters are documented in the schema (e.g., rate as decimal, annualDebtService as annual). The description adds context that all parameters are optional and that metrics are computed from whatever is supplied, which is a valuable semantic beyond the schema. It also explicitly says each metric is returned only when inputs are present, guiding parameter usage. This compensates for the 40% of parameters that lack schema descriptions, though the description does not enumerate specific parameter relationships (e.g., how NOI derives from revenue - expenses), which the schema partially does. Given the high parameter count and partial coverage, this is strong.

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 clearly states the tool computes a specific set of commercial real estate underwriting metrics from raw figures, naming each metric (NOI, cap rate, DSCR, etc.). It distinguishes itself from siblings by explicitly saying to use it instead of computing CRE ratios yourself, which differentiates it from tools like calculate_payment or size_loan.

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 explicitly instructs agents to supply whatever data is available and notes that each metric is returned only when its inputs are present, which guides when to use the tool (when raw figures are available) and implicitly when not (when you need a single payment calculation, use calculate_payment). It also says to use this instead of self-computation, providing clear usage direction.

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