PromptSpend

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

The description adds meaningful behavioral detail beyond the readOnly/openWorld annotations by naming history compounding, cache writes, long-context tiers, and reasoning tokens as modelled factors. However, with openWorldHint=true, it does not disclose where pricing data comes from, how fresh it is, or what the response shape looks like, leaving meaningful behavioral gaps.

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 tight sentences, front-loaded with what is computed, then the modelling factors, then the usage rationale. There is no filler, and every sentence contributes information.

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 10 parameters, a nested usage object, and no output schema, the description is reasonably complete on the computation model. Still, it omits currency, pricing data source/freshness under openWorldHint, and any indication of the return structure beyond 'on each model', which leaves some gaps for an agent reasoning about outputs.

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 100%, so parameter meanings are already well documented in the schema. The description only lightly clarifies the workload-vs-measured-usage distinction and does not add format or interpretation details beyond what the schema provides, making baseline 3 appropriate.

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 states a specific verb and resource: estimating monthly cost per model for either a described workload or measured usage. It does not explicitly differentiate itself from siblings get_price and find_cheaper, though the modelling behavior implies it is the calculation tool rather than a raw price lookup or optimization helper.

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 phrase 'Use this instead of multiplying prices by hand' gives implied usage guidance, but it does not state when to choose this over get_price or find_cheaper, nor does it give explicit preconditions or exclusions. The usage context is inferable but not fully specified.

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

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

Annotations already establish readOnlyHint=true and openWorldHint=true. The description adds useful behavioral context by clarifying that results are candidates for testing rather than a recommendation, but it does not disclose ranking, limitations, or output shape.

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?

Two tightly written sentences, with the core purpose front-loaded and the important caveat immediately following. There is no 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?

For a seven-parameter retrieval tool with no output schema, the description gives only a high-level indication of what is returned. It is sufficient to understand the tool's intent, but it does not describe the candidate result structure or any workload assumptions beyond the schema defaults.

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 100%, so all seven parameters are already documented, including defaults and the min_capability behavior. The description reinforces the capability threshold and workload basis but adds no parameter-specific semantics beyond 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?

The description states a specific task: finding cheaper models that meet or exceed the given model's capability, costed on the same workload. It also clarifies the return is test candidates, not a recommendation, which distinguishes it from the price/cost 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?

The intended use is implied by "cheaper models that score at or above the given one," but there is no explicit when-to-use or when-not-to-use guidance relative to get_price or estimate_cost. The caveat "Returns candidates to test, not a recommendation" helps interpretation but not tool selection.

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

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

Annotations already declare readOnlyHint and openWorldHint, so the safety profile is covered; the description adds genuinely new behavior by disclosing that disputed=true signals source disagreement and that close matches are returned for unknown models. It doesn't say whether an unknown model returns an error or empty results beyond the close-match hint, but the key edge cases are surfaced.

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 tight sentences: what it returns, the disputed/close-match edge cases, then the sibling routing. Nothing is repeated and the essential facts are front-loaded.

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 return-shape burden and does so for the main fields (price, source, last-confirmed date, disputed flag, close matches). The only gap is that it never positions the tool against find_cheaper, leaving one sibling relationship unexplained.

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 both model_name and the past-date as_of are already documented in the schema. The description reinforces the single-model scope and USD-per-million-token basis but adds no format or syntax detail for as_of beyond what the schema states, so the baseline 3 applies.

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 ("list price for one LLM") plus the unit semantics (USD per million tokens), which an agent can act on immediately. It also tells the agent what the payload includes (source, last-confirmed date), distinguishing it from siblings that aggregate or compare prices.

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?

Explicitly routes workload costing to estimate_cost, which is the most likely confusion given the sibling set. It does not mention find_cheaper at all, so the agent gets no guidance on when to look up a single price versus search for a cheaper alternative.

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