UMich Dining

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

With only readOnlyHint/openWorldHint in annotations, the description carries real behavioral weight: it discloses that numeric caps silently drop dishes lacking published values, that allergen flags are unreliable and indistinguishable between 'contains' and 'may contain', and that the gluten-free trait is not coeliac-safe. This is exactly the kind of caveat an agent needs before trusting results.

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?

Filter semantics are front-loaded and each sentence carries substantive information, including the safety caveats. It is dense but not padded; the dedicated gluten-free pantry list is the only line that feels more site-trivia than tool guidance.

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?

An output schema exists so return values need not be described, and the filter semantics plus data-quality caveats make the tool safe to invoke. The gaps are minor: no guidance on hall/limit or on choosing this over search_dishes.

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 only 11%, so the description must compensate, and it does for six of nine params: exclude_allergens (OR semantics), require_traits (AND semantics), min_mhealthy (tier threshold with an example), max_carbon, max_calories and max_sodium_mg. date/hall/limit are left to the schema, which covers date but not hall or limit.

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 ('Filter the dining hall menus') plus the exact dimensions being filtered (allergen, trait, nutrient density, carbon, calories). It never differentiates itself from the sibling search_dishes, which is a likely-overlapping verb, so it falls short of a 5.

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 what each filter does but gives no when-to-use guidance and never names or contrasts itself with the alternatives (search_dishes, get_menu). An agent must infer from the sibling names which tool to pick.

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=true and openWorldHint=true, so the safety and data-scope profile is covered. The description adds that both open and closed halls are returned along with today's service periods, which is useful framing, but it discloses no freshness/timezone caveats for a tool whose whole value is 'now'.

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 sentence that front-loads the resource and the temporal scope, then covers the full result shape (serving, not serving, today's periods). No filler or redundancy.

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 parameters and an output schema present, the description does not need to explain return values, and it correctly stays high-level. The only shortfall is not distinguishing itself from get_hours, which an agent in this sibling set would likely want clarified.

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 tool takes zero parameters, so there is no parameter semantics to document; the baseline for a parameterless tool is 4. Nothing in the description contradicts or is needed for the empty 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 (return) and resource (dining halls) with a clear temporal scope (at the current time) and distinguishes the output into currently-serving vs. not-serving. It never names the obvious sibling get_hours, so the boundary between 'open now' and 'hours' must be inferred.

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 'current time' framing implies when to use it (a point-in-time open/closed question), but there is no explicit statement of when to prefer this over get_hours for a schedule or list_locations for a directory. Usage guidance is inferable but not stated.

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=true and openWorldHint=true, so safety is covered. The description adds real behavior beyond that: the response combines published service periods for one day with the hall's recurring weekly schedule, and an omitted hall fans out to all halls. Error/empty cases (closed days, out-of-window dates) are still not addressed.

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 tight sentences with zero filler. The core behavior is front-loaded and the optional-argument default follows immediately as the one piece of operational nuance worth stating.

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 an output schema present, the description need not explain return values, and annotations cover the read-only/open-world profile. It covers the multi-hall default and the composed return content, though it leaves the edge case of dates outside the publication window to the schema.

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 50%: the date parameter is well documented in the schema (format plus the today-3 to today+14 publication window), while hall has only an enum and no description. The description compensates for the hall gap by stating it is optional and that omitting it returns all halls, but adds nothing about date semantics.

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 gives a specific verb and resource: it returns the service periods published for a dining hall on a requested day, plus that hall's regular weekly schedule. That is clearly distinct from get_menu (dishes) and find_open_now (live status), though it never names a sibling to make the distinction explicit.

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 hints at usage by noting that omitting the hall argument returns every hall, which tells the agent how to broaden the query. But there is no guidance on when to prefer this tool over find_open_now or get_menu, and no stated conditions or exclusions.

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 cover safety (readOnlyHint) and scope (openWorldHint), so the bar is lower, but the description goes further by disclosing meaningful lookup behavior: case-insensitive matching, exact-match priority, unique-substring fallback, and that ambiguous names are reported rather than guessed. That is real behavioral context not present in the schema. It stops short of describing not-found behavior or date defaulting.

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, no filler, with the core capability stated first and the matching rules second. The list of nutrition fields is slightly enumerative but earns its place by concretely defining 'full nutrition panel'. Front-loading is good.

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 an output schema present, return values need not be described, and the description covers the tricky input semantics (naming/ambiguity) that an agent could otherwise get wrong. It is still silent on what happens when no dish matches and on what the omitted 'date' parameter defaults to, which the schema hints at but does not resolve.

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 adds genuine meaning for the 'item' parameter by explaining the matching algorithm and its precedence rules, which the schema's terse 'The dish name' does not convey. Hall and date semantics remain fully delegated to 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 gives a specific verb and resource ('Get the full nutrition panel of one dish') and enumerates exactly what the panel contains (serving size, calories, fat, cholesterol, sodium, carbohydrate, protein, micronutrients). This clearly separates it from siblings like get_menu and search_dishes, which are about discovery rather than per-dish nutrient detail.

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: the caller must already know a dish name and want its nutrition rather than the menu listing. There is no explicit statement of when to prefer this over get_menu or search_dishes, nor any prerequisite guidance. The alternative tools are never named.

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 and openWorldHint, so the safety profile is covered. The description adds genuinely useful operational context: the default date behavior and the publication window of today-3 to today+14, which tells an agent which requests will fail. It does not mention pagination or payload size, keeping it from a 5.

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 packed sentences, front-loaded with the core action and followed by the date constraint. The parenthetical field list is dense but earns its place by telling the agent what comes back. No filler or restated boilerplate.

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 an output schema present, the description need not explain return shape, yet it helpfully summarizes it. Annotations cover the read-only/open-world profile and the date window is spelled out. The only real gap is the unspecified 'meal' parameter, which an agent must infer from the enum.

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 67%. The description reinforces the date parameter's semantics (default today, valid window) and implies meal coverage, but the 'meal' enum has no schema description and receives no explanation in the description either, so one parameter remains undocumented across both surfaces.

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 ('Get the menu of one dining hall') and enumerates the payload in useful detail (meals, stations, dishes, traits, allergens, nutrition). It implicitly separates itself from search_dishes/filter_dishes by scoping to a single hall, but never names an alternative, so it stops short of full sibling differentiation.

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?

Gives a clear context cue (single hall, date defaults to today) and a real constraint (menus exist only from today-3 to today+14), but offers no when-not guidance and never points to search_dishes or filter_dishes for cross-hall or criteria-based lookups. Usage is implied rather than prescribed.

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 description's added value is the fixed set of ten locations, the per-day service-period behavior, and the live 'serving right now' flag — all useful context beyond what the annotations state. It doesn't address auth, caching, or what happens for a date outside the publish window, which the schema hints at but the description does not.

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?

Opens with the verb and resource, then the field list, then the exclusion sentence — a logical front-loaded order with no filler. The returned-field enumeration is long but each item is meaningful for an agent deciding whether the response is sufficient.

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?

A read-only list tool with one optional, fully documented parameter, an output schema, and a description that covers scope, returned fields, and the live-status nuance. An agent has everything needed to call it correctly and know when the result is insufficient.

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% with a documented YYYY-MM-DD pattern and a stated publish window (today-3 to today+14), so the parameter is fully specified structurally. The description only alludes to the date via 'requested day' and adds no format or range detail beyond the schema, which is the baseline case.

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 (List) and a precisely bounded resource (the ten U-M dining halls), then enumerates the fields returned. The scope sentence excluding cafes, markets, bakeries, grills, prices and meal plans makes it unmistakable which locations are covered.

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 out-of-scope list clearly tells the agent when NOT to use this tool, and 'the service periods published for the requested day' implies the date-scoped use. However, it never names the sibling that handles the excluded categories (e.g. get_menu for menus, filter_dishes for prices), so routing still requires inference.

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 safety is covered. The description adds genuinely useful behavioral semantics beyond that: matching is a case-insensitive substring on dish names, which tells the agent partial queries like 'pizza' will match 'pepperoni pizza slice'. It stops short of describing result ordering, default limit, or pagination.

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 tight sentences with no filler. Scope is front-loaded and the matching rule follows immediately, so nothing needs to be re-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?

An output schema exists, so return values need not be explained, and annotations cover the safety profile. What remains missing is any signal separating this from filter_dishes and any note on default result count or ordering for a tool with a limit parameter.

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 75% (date, hall, and query are documented; limit is not), so the schema carries most of the load and a baseline of 3 applies. The description adds match semantics for query and reinforces hall scoping, but says nothing about date or limit behavior that would compensate for the remaining gap.

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 ('Search dish names') plus scope ('across the dining halls, or within one hall'), so the agent knows exactly what is being queried. It does not distinguish itself from the sibling filter_dishes, which is the tool an agent is most likely to confuse it with.

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: the 'or within one hall' phrasing hints that the hall parameter narrows scope, but there is no explicit when-to-use statement and no mention of filter_dishes or get_menu as alternatives. An agent must infer the routing decision from the tool name alone.

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