Tower and Wing: FHA-insured hospital loans (HUD Section 242)

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

Annotations already declare readOnly, idempotent, non-destructive and closed-world, so the safety profile is covered. The description adds genuinely non-obvious behavior: the response bundles every active loan with a detailed field list, plus totals and earlier terminated FHA hospital loans in the same city from HUD's terminated file — an unexpected inclusion an agent could not predict.

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?

A single front-loaded sentence with the resource and lookup keys stated first, then the payload. The long parenthetical field list is dense but earns its place given there is no output schema; it borders on run-on but contains 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?

With no output schema, the description does the necessary work of enumerating returned fields, and it discloses the non-obvious terminated-loan side data. It omits failure/not-found behavior and how name collisions are resolved, which are the remaining gaps for a one-parameter lookup tool.

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% and the schema already gives the same three examples (name, slug, 8-digit FHA number), so the description largely restates structured data. It reinforces that all three forms are accepted by the single parameter, but adds no syntax or disambiguation rules beyond that.

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 resource ('one hospital or health system with an active FHA-insured loan') and the three accepted identifier forms, and it enumerates what the response contains. The singular lookup scope cleanly separates it from the plural sibling search_hospital_loans and from get_portfolio_stats / get_program_rules.

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 only implied: an agent can infer this is the exact-identifier lookup rather than a search, but the description never states when to use this versus search_hospital_loans or what to do when the identifier is ambiguous. No exclusions or prerequisites are given.

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, closed-world and non-destructive, so the safety bar is low. The description adds genuine value beyond them: the data provenance ('computed from HUD's files') signals derived/snapshot figures, and it specifies historical coverage back to 1968 including terminations and insurance claims, which the annotations cannot convey.

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?

A single dense sentence with no filler; the core purpose and scope are front-loaded before the metric enumeration. The long list of metrics is information-dense rather than wasteful, though it borders on being a run-on read.

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 compensates by enumerating what the response contains (active loans, balances, median note rate, program counts, construction, bond loans, and the since-1968 history). For a read-only stats tool with one fully-documented param and complete annotations, this is close to sufficient; only freshness/as-of dating is left implicit.

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?

The single parameter is fully documented in the schema (two-letter code, omit for whole US), so the schema carries the syntax. The description reinforces the semantics by framing it as 'nationally or for one state,' which confirms the omission behavior, though it adds no format detail 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?

States a specific verb+resource (aggregate totals for FHA-insured hospital loans) and enumerates the metrics returned, plus the national-vs-single-state scope. It implicitly separates itself from search_hospital_loans (row-level search) and get_hospital (single record) by being a totals/computation tool, but never names those siblings, so differentiation is inferred rather than stated.

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 makes clear you call it for aggregate figures and can scope by state or omit for the whole US, which implies the usage context. However, it gives no explicit when-to-use-this-instead-of guidance against search_hospital_loans or get_hospital, and no exclusions or prerequisites.

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, closed-world and non-destructive, so the safety profile is covered. The description adds genuine beyond-schema context: the content is 'as hospital.loans publishes them, with citations,' which tells the agent the authority and traceability of the answer. It stops short of stating freshness/version or that the content is static reference material.

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?

A single dense sentence that front-loads the source and scope, then lists the four coverage areas in the same order as the topic enum. Every clause maps to a queryable topic; it is long but not padded.

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 describing returns, and it does so thoroughly for all four topic values. Combined with a fully documented two-parameter schema and complete annotations, an agent has everything needed to call this correctly and know what to expect.

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% and both enums are already documented in the schema, so the baseline is 3. The description still adds value by spelling out the substance behind each topic (the four programs with limit/term/MIP, the 24 CFR tests, fees/reserves, process days) and by confirming the program parameter's role in narrowing topic=programs.

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 resource (HUD hospital mortgage insurance rules) and enumerates exactly what it covers: the four programs, Section 242 eligibility tests, fees/reserves, and process steps. It is distinguishable from the siblings (get_hospital, search_hospital_loans, get_portfolio_stats) by being a rules reference rather than a data lookup, though it never says so explicitly.

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 only implied: an agent can infer this is the tool for 'what are the rules' questions versus loan/hospital data tools. There is no explicit when-to-use, when-not-to-use, or named alternative, so nothing actively routes the agent away from a sibling.

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 and closed-world, so the safety profile is covered; the description adds real context on top: the data reaches back to 1968, terminated loans and insurance-claim endings are included, and the result payload fields are named. It omits that status defaults to 'active' (so an agent may not realize terminated loans require an explicit parameter) and says nothing about the 20-row cap or the 1000-row paging ceiling.

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 sentences, front-loaded with the verb and data scope before the filter and return lists. The return-field roll-call is slightly long but justified because there is no output schema to carry it; nothing 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?

With no output schema, the description takes on the return-value burden and does it — FHA number, hospital, program, amount, note rate, dates, balance or termination outcome, plus the hospital page link. Combined with fully documented schema parameters, an agent has everything needed to call this 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% and the schema already explains status enum values, program meanings, sort, limit and offset, so the baseline is 3. The description restates the filter dimensions and adds the note that status can include loans that ended in an insurance claim, but contributes no syntax, format or default information beyond what the schema carries.

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?

Opens with a specific verb and resource ('Searches every FHA-insured hospital loan in HUD's files') and immediately bounds scope by naming the four programs and the active/terminated split. That makes it easy to separate from get_hospital (single-entity lookup) and get_portfolio_stats (aggregates) without opening a schema.

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 enumeration of filter dimensions (hospital, system, city, state, program, status, endorsement years) tells the agent what queries this tool answers, and the 'active loans or terminated ones since 1968' phrasing establishes the temporal reach. It stops short of explicit routing guidance — it never says to use get_hospital for a single facility or get_portfolio_stats for rollups — so it is clear context without exclusions.

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