Bedcount: student housing data by campus

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

Annotations already declare this as a read-only, idempotent, non-destructive, closed-world operation, so the safety profile is covered. The description adds useful behavioral context beyond annotations: data source (federal IPEDS), vintage (fall 2024, 2024-25), population restriction (3,000+ full-time students), and the specific metrics returned.

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 front-loaded with the core purpose and uses only two sentences, with no filler. However, the first sentence is a long, dense enumeration of data fields, which is useful but makes the definition harder to scan quickly than a more structured presentation would.

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, so the description carries the burden of explaining return values, and it does so comprehensively by listing enrollment, housing beds, bed gap, charges, aid budgets, and the Fannie Mae test. Combined with fully described input parameters and annotations covering safety, an agent has enough information to call the tool correctly.

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 three parameters are already fully documented. The description repeats the lookup methods (name with optional state, or IPEDS UNITID) but does not add syntax or selection semantics beyond what the schema already provides, such as the explicit 'unitid overrides name' behavior.

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/resource: retrieving one US four-year campus from IPEDS data, with a clear scope of 3,000+ full-time students. It distinguishes the tool from a general campus search by emphasizing 'One' campus and specifying exact lookup methods. The extensive data fields further clarify what the campus record contains.

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 explains how to invoke the tool: by name with optional state, or by IPEDS UNITID. This gives clear context for lookup, but it does not explicitly contrast with the sibling search_campuses or state when not to use this tool, such as when the campus is unknown.

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 readOnly, idempotent, non-destructive, closed-world, so the safety profile is covered. The description adds genuinely non-obvious context: a hidden base cohort (US four-year, 3,000+ full-time students) that constrains every query, plus exactly what each returned row contains. It does not address pagination limits or what happens when filters yield nothing, but it meaningfully exceeds the annotation baseline.

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 dense sentences: the first front-loads the population, filters and ranking vocabulary; the second enumerates returned fields. Every clause carries information, though the long comma-chained filter list is heavy and sacrifices scannability for completeness.

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 11 optional parameters and no output schema, the description supplies the missing return-value detail (fall 2024 enrollment, beds, bed gap, housing charge, campus link) and the implicit cohort floor. The remaining gap is the tension between the stated 3,000+ floor and a min_fulltime minimum of 0, which is never reconciled.

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 baseline is 3, but the description adds a semantic constraint absent from the schema: the population is pre-filtered to 3,000+ full-time students at four-year US institutions, which interacts with the min_fulltime parameter (min 0) and prevents misreading that parameter as removing the cohort floor. It also documents the output row fields, which no output schema provides.

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 (Lists) and a well-scoped resource: US four-year campuses with 3,000+ full-time students, enumerating both filter dimensions and ranking dimensions. The listing behavior is clearly distinct from get_campus's single-record lookup, but the description never names that sibling, so differentiation is left to inference.

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?

Usage is implied by the filter and ranking vocabulary — an agent can infer this is the tool for discovering/ranking many campuses, with get_campus for one. There is no explicit 'use this when' statement, no mention of alternatives such as get_campus, and no guidance on defaults (sort=bed_gap) or when a filter combination is meaningful.

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, idempotentHint and destructiveHint=false, so safety is covered. The description goes beyond them by disclosing the computation logic — that the maximum loan is the smallest of the LTV, DSCR and debt-yield tests and which test binds — which is meaningful behavioral context for interpreting results. It doesn't cover error behavior or unit handling, keeping it at 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?

Purpose is front-loaded in the first clause, but the rest is a single run-on sentence that strings inputs, outputs, override logic and preset provenance together without punctuation breaks. It is dense but readable; splitting the return-value clause out would help scanning.

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 14 parameters and no output schema, the description's enumeration of return values (NOI, max loan, binding test, debt service, loan per bed) fills the gap the missing output schema leaves. Preset override semantics are implied through the schema rather than stated in the description, a minor shortfall.

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 the baseline is 3; the description earns the bump by naming the input families (beds, rent per bed, months collected, vacancy, expenses, reserves, value, own interest rate) and flagging the debt yield as optional and the rate as 'your own,' i.e. an override rather than a preset value.

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 names a specific verb and resource ('by-the-bed loan sizer') and enumerates both the input drivers and the outputs (NOI, max loan, binding test, annual debt service, loan per bed). Siblings (get_campus, search_campuses) are unrelated lookups, so no differentiation burden exists, but the scope statement is precise regardless.

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?

It implies usage context by calling itself a loan sizer and by explaining that presets are 'Freddie Mac's published student housing LTV/DSCR tiers,' and it notes min_debt_yield is optional. It never states when to use this vs. an alternative, nor any precondition such as needing a property value or a quoted rate before sizing.

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