Human-verified directory of non-alcoholic bars: 1,500+ venues in 70 cities, rated and re-checked.
Scored across 5 tools
Each tool has a largely distinct role, but find_drink (drink-first discovery) and search_venues (venue-first discovery) overlap in that both return venues, and get_menu vs get_venue both fetch by slug. Descriptions do clarify the difference (drink query vs venue filters; menu items vs venue record), so misselection risk is low but present.
All five tools follow a clean verb_noun convention (find_drink, get_menu, get_venue, list_cities, search_venues). No mixing of camelCase, no vague verbs, fully predictable.
Five tools is well-scoped for a venue-directory server: two discovery paths, two detail lookups, and one list. Slightly lean—there is no dedicated brand lookup despite brand data being surfaced—but nothing feels redundant.
The surface covers discovery (find_drink, search_venues), detail (get_venue, get_menu), and city coverage (list_cities), with full read coverage of the domain. Minor gaps: no brand-level lookup (brandSlugs are returned but not searchable) and read-only by design, but no dead ends for core workflows.
Find venues whose VERIFIED menu lists a specific non-alcoholic drink — e.g. a real NA negroni, non-alcoholic IPA, zero-proof espresso martini, NA spritz or kava serve — optionally in one city. Every result comes from the venue's own published menu (read verbatim and dated), never from reviews or guesses, so a listed venue genuinely pours that drink. Returns the matching menu items (name, description, price where printed, on-draft flag), the menu's last-confirmed date, and the canonical nabarfinder.com URLs for citing. Drinks currently on menus (24 kinds across 644 venue-pours): na-ipa (125), na-lager (84), na-spritz (83), na-sparkling-wine (70), na-negroni (59), na-white-wine (29), kava-serve (24), na-mocktail-sour (22), na-mojito (20), na-margarita (19), na-espresso-martini (18), na-red-wine (18), na-stout (11), na-old-fashioned (10), na-paloma (10), na-gin-and-tonic (7), na-margarita-spicy (7), functional-adaptogen-cocktail (6), na-aperol-spritz (5), na-moscow-mule (4), na-negroni-sbagliato (4), na-sour (4), na-highball (3), na-wheat-beer (2). Read-only.
| Name | Required | Description | Default |
|---|---|---|---|
| city | No | Limit to one covered city — slug ('austin-tx'), name ('Austin') or 'Austin, TX'. Omit for every city. | |
| drink | Yes | The drink, as a slug ('na-negroni') or plain words ('non-alcoholic negroni', 'NA IPA', 'virgin mojito'). Unknown drinks return the catalog of known slugs. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full burden and does so: it promises results come 'read verbatim and dated', 'never from reviews or guesses', and ends with 'Read-only'. It also discloses the returned freshness signal, the menu's last-confirmed date, and canonical citation URLs.
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 core purpose is front-loaded in the first sentence, followed by verification, return values, and accepted drink slugs in a logical order. The catalog is long and includes arguably extraneous pour counts, but it is structured and each part serves selection or invocation.
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 two simple parameters and no output schema, the description explains all returned fields (menu items, on-draft flag, last-confirmed date, URLs) and lists accepted drink values. Nothing needed to call the tool correctly is missing.
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 input schema already covers both parameters at 100%, and the description adds a current catalog of 24 valid drink slugs and counts, going beyond the schema examples. It does not repeat schema syntax, and the catalog meaningfully helps an agent pick a valid drink parameter.
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 opens with a specific verb-object: 'Find venues whose VERIFIED menu lists a specific non-alcoholic drink', and it names concrete beverage examples. This clearly differentiates the drink-first venue lookup from sibling tools such as get_venue or get_menu.
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 'optionally in one city' and 'omit for every city' guidance makes the scope clear, and the verified-menu emphasis implies this is the tool for drink-specific discovery. However, it never names search_venues or get_menu as alternatives or states when NOT to use this tool.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Fetch the full verified record for one NABarFinder venue by its stable slug: city, neighborhood, address and coordinates (when it has a fixed location), venue type, naProgramStrength (1–5), full description, NA brands stocked (brandSlugs), last-verified date, canonical nabarfinder.com URL, and — when the registry has them — website, phone, hours, signature drinks, and a sourced press quote. Fields absent from the registry are omitted, never invented. Read-only.
| Name | Required | Description | Default |
|---|---|---|---|
| slug | Yes | The venue's stable slug id, e.g. 'sans-bar-austin' — as returned by search_venues or the /api/v1/venues.json feed. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden, and it delivers: it explicitly states that fields absent from the registry are omitted and never invented, and that the operation is read-only. This goes well beyond the schema by setting accurate expectations about absent data and data integrity.
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 a single, front-loaded sentence that immediately names the action and resource, then lists the returned fields. It is dense but every clause adds useful information, and because there is no output schema, the length is justified.
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 tool has only one parameter, no annotations, and no output schema, the description is remarkably complete: it enumerates the full set of return fields, explains omission behavior, and declares the read-only nature. An agent can reliably decide to invoke it and interpret its results without needing additional context.
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 input schema already fully documents the lone slug parameter with an example and its source ('as returned by search_venues or the /api/v1/venues.json feed'), so schema coverage is 100%. The tool description repeats 'stable slug' but adds little parametric meaning beyond what the schema already provides, so the baseline score of 3 is 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 opens with a specific verb and resource: 'Fetch the full verified record for one NABarFinder venue by its stable slug.' It enumerates the exact fields returned and references search_venues as the source of the slug, which clearly distinguishes this single-venue lookup from sibling tools like search_venues and list_cities.
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 gives clear context: this is a single-venue lookup keyed by a stable slug, and it tells the agent the slug comes from search_venues or the /api/v1/venues.json feed. It implies when to use it, but it does not explicitly state exclusions or alternatives such as 'use search_venues when you need a list.'
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
List every city NABarFinder covers, each with its state, number of human-verified venues, city slug (for search_venues), and canonical nabarfinder.com city guide URL. Takes no input. Read-only.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are present, so the description carries the full burden. It explicitly states 'Read-only' and 'Takes no input', which are key behavioral details. It does not mention potential side effects or limitations, but the read-only nature makes such omissions acceptable.
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 a single, concise sentence that efficiently communicates the purpose, output content, and behavior (read-only, no input). It is well-structured with no 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?
For a simple, parameterless list operation, the description is complete. It covers the purpose, output fields, and behavioral constraints (read-only), making it sufficient for an agent to decide when and how to invoke the 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?
The tool has no parameters, so schema coverage is effectively 100% (nothing to cover). The baseline is 3, and the description adds no parameter-specific information because none exist.
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's function (list all cities NABarFinder covers) and specifies the output fields (state, number of human-verified venues, city slug, canonical URL). It distinguishes from sibling tools by focusing on cities rather than individual venues or searches.
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 implies when to use it (when a full list of cities is needed) and notes it takes no input, but it does not explicitly contrast with the sibling tools or state an alternative for filtering. The implied usage is clear enough.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Search NABarFinder — a human-verified directory of 1548 venues with real non-alcoholic (NA) drink programs across 70 cities in the US, UK, and beyond. All filters are optional and combine with AND; with no filters it returns the top-rated venues overall. Returns at most 25 venues (sorted by NA Program Strength descending, then name), each with name, slug, city, venue type, naProgramStrength (1–5; 5 = fully alcohol-free or a dedicated NA program, 1 = a few NA options), a one-line summary, and the canonical nabarfinder.com page URL for citing/linking. The response includes the total match count; if it exceeds the cap, narrow with filters. Use get_venue with a result's slug for the full record. Read-only.
| Name | Required | Description | Default |
|---|---|---|---|
| city | No | Limit to one covered city. Accepts a city slug ('austin-tx'), a plain name ('Austin'), or 'Austin, TX'. Call list_cities for the covered set — cities outside it return zero matches. | |
| query | No | Free-text filter over venue name, neighborhood, venue type, and description. Case-insensitive; every word must match (e.g. 'kava lounge', 'zero proof cocktail'). | |
| min_program_strength | No | Minimum NA Program Strength on the registry's 1–5 scale (5 = fully alcohol-free / dedicated NA program). E.g. 4 returns only the strongest programs. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden and delivers: read-only status, the hard 25-result cap, sort order, the presence of a total match count, and the exact per-result fields. It also warns that cities outside the covered set return zero matches.
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?
Front-loaded with identity and scope, and every sentence carries information (cap, ordering, fields, citations, routing). It is delivered as one dense block rather than scannable segments, which slightly hurts structure but not content economy.
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 must describe returns — and it does, naming each field, the cap, ordering, and the total match count. Combined with the zero-required-parameter invocation model and sibling routing, an agent has everything needed to call and interpret this 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%, so the schema already documents city slug/name formats, the free-text AND-matching behavior, and the 1–5 strength scale. The description largely restates the strength scale meaning rather than adding new parameter-level guidance, 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 (search a human-verified directory of 1548 NA venues) plus scope (70 cities across US/UK) and result semantics (max 25, sorted by NA Program Strength). It explicitly distinguishes itself from siblings by routing full records to get_venue and the covered city set to list_cities.
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 explicit conditions: all filters are optional, combine with AND, no filters returns top-rated overall, and when the total match count exceeds the cap the agent should narrow with filters. It also names the follow-up tools (get_venue by slug, list_cities for the covered set) with the reason to use each.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Editorial guide to restaurants and bars — hand-picked venues with opinions, not a places database
Search nearly 20,000 verified restaurants and bars across 24 cities plus the New Mexico region, with venue detail and curated lists.
Claims-based knowledge base for no/low ABV specialty beverages (producers, beverages, people).
Find places by what you want (work friendly, gluten free, great coffee), scored from real reviews.
Add one secure layer between your agents and this server.
