IMPOSSIBREW

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

Annotations already declare readOnly/openWorld/idempotent, so the safety profile is covered. The description adds genuine behavioral value the annotations cannot: an authoritative source ('same source as the AI Brewer'), a strict output constraint (quote facts only) and explicit prohibitions on health-effect and medical-advice content. It does not mention coverage limits or fallback behaviour when a topic has no matching fact.

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 core action and source, followed by the topic list and constraints. The inline topic glosses are dense but each earns its place; the parenthetical groupings are slightly list-heavy but aid 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?

For a one-parameter, read-only lookup with no output schema, the description covers source authority, topic scope and quoting constraints well. It does not describe the return shape (facts plus citation?) or what happens when a topic returns nothing, which is the main residual gap.

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 0%, but the single parameter is a self-documenting enum. The description goes beyond the bare enum labels by glossing what each topic covers — dietary (gluten, vegan, allergens, carbs, sugar), ingredients (Social Blend, ABV), safety (pregnancy, medication, driving, daily limit) — which materially helps topic selection.

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 ('Answer customer questions from the official IMPOSSIBREW policy') and enumerates the twelve supported topics, which clearly separates it from browse_beers, get_product and order_help. An agent can route on this without opening the 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 topic enumeration plus 'Quote the returned facts only; never add health effects, gluten-removal methods or medical advice' defines the domain and the negative space an agent must respect. It gives clear context but stops short of explicitly naming the sibling alternative (e.g. order_help or product_mentions) for out-of-scope questions.

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 readOnly, idempotent and non-destructive, so the main added value is the fixed cardinality (exactly six results) and the fact that prices are current and per-pack — both real behavioral traits not derivable from the annotations. It stops short of saying how availability is represented or whether results are cached.

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, both earning their place: the first delivers content and return shape, the second bounds scope. Front-loaded with the verb and resource, with 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?

The description usefully covers the return shape (six beers with prices and availability) in the absence of an output schema, which is a real contribution. But for a tool with an undocumented free-text filter, leaving "query" completely unexplained means an agent lacks enough to invoke it well.

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 0% for the single optional "query" parameter, so the description carries the full burden — yet it never mentions what query accepts or how it filters the six beers (name, style, price?). This is the definition's clearest gap: the agent could call it with no query and never know what filtering was possible.

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 ("Read") and resource ("six featured IMPOSSIBREW beers") with the returned attributes (exact-pack prices and availability) spelled out. The closing exclusion ("Not a full store catalogue") bounds scope, but it never names the sibling (e.g. get_product or compare_beers) that handles the broader catalogue, so differentiation is only partial.

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?

"Not a full store catalogue" implies a when-not-to-use condition, which is more than most definitions offer. However, no alternative tool is named for the full-catalogue or comparison cases, and there is no guidance on when to prefer this over compare_beers or get_product, so the agent must infer the routing.

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?

Goes well past the annotations: it declares that no order is placed and no payment is taken (consistent with readOnlyHint=true), that prices/gifts come from Shopify live readback and must be quoted not estimated, that subscriptions are cancel-any-time, and it encodes the exact gift tiers by box count. It also flags a mid-flow signal (need_shirt_size) the agent must react to.

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 and the no-payment caveat are front-loaded, then gift tiers and the shirt-size conditional follow in a logical order. Some phrases are redundant ("quote them exactly, never estimate" plus the readback sentence), but the density is justified for a commerce flow with this much policy.

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 work of explaining returns: a checkout link, plus the subscription-equivalent beers and savings for one-time baskets. The remaining gap is the items/packCans semantics, which the agent needs to build a valid basket.

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 0%, so the description carries the burden. It gives real meaning to frequency (subscription cadence) and shirtSize (S-4XL, only when need_shirt_size is returned), but says nothing about the items array — productId enum values, the quantity max of 6, or what packCans means — leaving the most important parameter under-explained.

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 — builds a real IMPOSSIBREW Shopify basket and returns a checkout link — with concrete scope (one-time vs subscription, impossibrew.co.uk checkout). It does not name the near-identical siblings build_box and build_shop_link, so the agent must infer the split from those tools' own descriptions.

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?

"Use whenever the customer wants to buy, order, try or subscribe" gives a clear positive trigger, and the subscription/one-time branch plus the need_shirt_size condition are spelled out. No exclusion or alternative routing is offered despite build_box, build_shop_link and open_shop sitting in the same namespace.

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?

Beyond the annotations, it discloses that pricing/availability are live and what the output contains (exact lines, subtotal, cart link), and it draws a firm boundary that no order or payment occurs. Some of that overlaps the readOnlyHint=true annotation, but the checkout-handoff detail is genuinely additive information.

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, front-loaded with the core action and inputs, followed by scope exclusions. Every clause carries information; 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 four optional parameters, no output schema and only hints in annotations, the description does enough: it explains inputs, the live-data basis, the return shape, and the no-side-effects boundary. It omits edge-case behavior (e.g. when a budget cannot be met, or what happens with zero constraints).

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 0%, so the description carries the full burden — and it names all four inputs with meaning ('specific beers, gluten-free only, number of cans and/or budget in GBP'), including the 'and/or' semantics that at least one constraint is expected. It does not mention ranges or the fixed beer-style list, which the schema does enforce.

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 uses a specific verb and resource ('Build a ready-to-review one-time IMPOSSIBREW basket') and enumerates the constraint inputs that shape it. It stops short of distinguishing itself from close siblings like build_basket or build_shop_link, so an agent must still infer which builder to pick.

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 gives clear usage context (constraints in: beers, gluten-free, cans, budget; live prices) plus explicit exclusions: 'Does not create an order, take payment or start a subscription; checkout happens on impossibrew.co.uk.' No sibling tool is named as the alternative, which is the only real gap.

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?

The negative disclosure — "it does not create a cart, order, payment or subscription" — precisely matches readOnlyHint=true/destructiveHint=false and tells the agent the call is side-effect free. It adds context annotations cannot express: shipping and discounts are resolved downstream at checkout, so the agent should not promise final pricing.

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?

Four sentences, front-loaded with the verb and scope, then constraints, then routing. Every clause carries a decision the agent would otherwise get wrong; nothing is repeated from the schema.

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 correctly states what comes back (a link only) and that the link is not a transaction. It does not say what the permalink looks like or how to recover when validation fails for a given variant, which would close the last gap for a tool that is one deep nested 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 0%, so the description must carry the load; it contributes the 1-to-12 line bound and the "one-time" framing, and implies variantId must be a currently valid variant. It says nothing about the productId enum, the 1-6 per-line quantity cap, or where to source variantId, so nested field meaning is left largely to the schema's structural constraints.

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?

"Construct an exact Shopify cart permalink for 1 to 12 one-time item lines" gives a specific verb plus resource, and the scope (permalink only, one-time lines) separates it cleanly from build_basket, build_box and open_shop. An agent knows this produces a URL, not a cart.

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?

Explicit exclusions with named alternatives: no recurring purchases, and route existing orders/subscriptions to order_help with a contact-page fallback if that tool is missing. The "after current variant/availability validation" prerequisite also tells the agent when in the flow to invoke it.

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 (readOnly, idempotent, non-destructive), so the description's burden is to add behavior. It does disclose that results are a 'source-backed position' rather than neutral fact, which is valuable and honest. However, 'refresh commerce through the product tools' is opaque jargon that signals some side-effect-adjacent intent without saying what it actually means, and nothing is said about coverage limits or what happens when an alternative has no published page.

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 useful part, so it is compact. But the third clause 'refresh commerce through the product tools' is vague filler that muddies an otherwise clear sentence, and the description has no structure separating what it does from what it returns.

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 one-parameter read tool with no output schema, the description is only partly sufficient. It never characterizes the return value (is it prose, a structured verdict, links to pages?) or the 3-item limit, though annotations do carry the safety profile.

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 0% for the single parameter. The description only says 'named alternatives', which is a restatement of the parameter name rather than an explanation. It never mentions the hard cap of 3 competitors or what an invalid/unpublished name yields, both of which an agent needs to construct a valid call.

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?

It states a verb and resource ('Read our canonical published comparison pages for named alternatives'), so an agent can tell it reads pre-written comparison content rather than computing a comparison. But it never distinguishes itself from the sibling tools compare_beers and compare_pack_value, which is exactly the confusion an agent would hit here, and the title 'Why IMPOSSIBREW?' adds no operational information.

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?

There is no statement of when to use this rather than compare_beers or compare_pack_value, and no prerequisites or exclusions. The only directional hint is an implicit positioning instruction ('distinguish it from independent proof'), which is about output framing, 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 readOnly, idempotent, non-destructive and openWorld, so the safety profile is covered. The description adds a genuine caveat that nutrition/allergen data stays per-label rather than being normalised across the comparison, but the wording ('remain specific to each label') is vague and no other behavioral detail (e.g. what happens with out-of-stock items) is given.

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 short sentences, zero waste, with the resource and the compared dimensions front-loaded before the caveat. Nothing could be removed without losing 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?

For a simple read-only comparison tool with no output schema, stating the compared dimensions effectively substitutes for return-value documentation. The one gap is that result shape and behaviour for unavailable beers are unstated, but overall it is sufficient to call 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?

There is a single 'ids' parameter with zero schema description coverage, but the enum of valid beer slugs is present in the schema. The description only restates the 2–4 count constraint that minItems/maxItems already enforce, and hints that ids are 'specific' beers — marginal added meaning.

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?

Uses a specific verb (Compare) on a specific resource (2–4 IMPOSSIBREW beers) and enumerates the comparison dimensions: flavour, pack prices, availability, source descriptions. This implicitly separates it from compare_pack_value (value only) and compare_alternatives, though no sibling is named 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?

The '2 to 4 specific ... beers' phrasing implies usage for a known, small set of products, which is a usable constraint, but there is no explicit when-not guidance and no routing to browse_beers or compare_alternatives. Usage must be inferred.

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 read-only, idempotent, open-world and non-destructive, so the description correctly focuses elsewhere. It adds real behavioral context: prices are live/fresh, ranking is constrained to identical beer and can size, and results exclude delivery, discounts and retailer prices with an explicit 'never promise the cheapest overall deal' caveat.

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, front-loaded with the action and metric, then the ranking rule and exclusions. No filler; the caveat sentence earns its place by preventing overclaiming.

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 convey the shape of the result (ranked per-can and per-litre values within comparable groups) and its limits. The main gap is the unaddressed input parameter, which is the only thing an agent needs to call it 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 0% and the tool takes one parameter (ids), which the description never mentions or explains. The enum contents in the schema imply valid beer keys, but nothing in the description tells an agent what to pass, whether it is required, or whether omission means 'compare all'.

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?

Names a specific verb and resource (compare one-time IMPOSSIBREW packs) and clarifies the metric used (fresh store prices, per-can and per-litre arithmetic). It does not explicitly distinguish itself from siblings like compare_beers or compare_alternatives, which is the only missing piece.

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 scope constraints (rank only same beer and can size; exclude delivery, checkout discounts, retailer prices) which imply when the tool applies. But it never names an alternative or a trigger condition versus compare_beers/compare_alternatives, so tool selection is left to 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 cover read-only, idempotent, open-world and non-destructive traits, and the description adds genuinely new context: retailer links are not live shelf-stock, delivery-coverage or price checks, and results carry 'source health'. That pre-empts the exact misinterpretation an agent would otherwise make; only the return shape and any staleness/ordering caveats are unstated.

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 short sentences, front-loaded with what is returned and followed by the critical 'not live stock' caveat. The wording is dense in places ('honest source health', 'separate online retailer buying destinations') but nothing is wasted.

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 one optional parameter, no output schema and full annotation coverage, the description gives enough to call it correctly and to interpret results cautiously. Only a hint of what a result entry contains (retailer name/link vs venue name/address) 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 single parameter has zero schema description coverage, so the description must carry the load, and it does: query is a town/postcode text filter that affects physical venues only, not retailer links. It omits the 80-character limit and matching behavior, but the semantics are clear.

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 the concrete resource: online retailer buying destinations plus verified UK physical venues, with source-health information. That distinguishes it from siblings like build_shop_link or open_shop, though the phrase 'separate online retailer buying destinations' is awkward and takes a second read to parse.

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 states the optional query filters physical venues only, and warns 'never infer availability', which is useful scoping. But it never says when to call this versus sibling tools such as build_shop_link or open_shop, leaving the choice to 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 cover safety (readOnly, idempotent, non-destructive), so the bar is lower; the description adds a genuine behavioral constraint — variant IDs must be taken from the response and never inferred, which implies a product-ID vs variant-ID distinction. It stops short of describing the shape of the returned facts or whether results can change between calls.

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 short sentences, zero filler, with the core action front-loaded and the critical ID caveat immediately after. Every clause carries information an agent needs.

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 single-parameter read tool with full annotation coverage and no output schema, the description supplies the action, the object, and the key identity caveat. What it omits — the specific fields in "product facts" and how it relates to browse/compare siblings — is minor but non-zero.

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 0% for the single required `id` parameter, and the description never names or explains it beyond implying it selects "one beer." It does add useful semantics by distinguishing the returned variant ID from the input identifier, but the enum values (lager, pale, tipa, lemon, grapefruit, golden) are left for the schema alone to convey.

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: read product facts and current exact variants for one beer. That clearly separates it from sibling tools like browse_beers or compare_beers, though it does not name those alternatives explicitly. Purpose is unambiguous but not sibling-routed.

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?

"Use the returned variant ID, never infer it" is guidance about consuming the result, not about when to call this tool instead of browse_beers, compare_beers, or product_mentions. Usage is implied (look up one beer's facts) with no exclusions or alternatives 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 declare readOnly, idempotent, openWorld and non-destructive, so the safety profile is covered. Beyond that the description adds real context: it renders an interactive UI, leads to a genuine Shopify basket, and imposes an explicit constraint not to inject custom product images because official pack shots are shown.

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 short sentences, front-loaded with what the tool opens and what the customer sees, followed by the outcome and the image constraint. Every sentence carries distinct information with no restatement of the name or title.

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 no output schema, the description carries the full burden and mostly meets it: it explains the rendered experience, the checkout outcome, and the image rule. It is slightly thin on what the agent receives back when the UI is opened (e.g., a widget handle vs. a plain confirmation).

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 the baseline is 4. The description reinforces this with "Accepts empty arguments," removing any ambiguity about whether arguments must be supplied.

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 ("Open the IMPOSSIBREW shop") and then describes exactly what the customer sees: a swipeable picker with pack shots, per-can prices and a gift ladder. That makes it distinguishable in spirit from browse_beers and build_basket, though no sibling is named 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 implied rather than stated: the customer picks boxes and taps Subscribe/Buy once to get a real Shopify basket. There is no explicit when-to-use vs. when-not, and the nearby tools (browse_beers, build_basket, build_box, build_shop_link) are never referenced as alternatives.

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 declare destructiveHint=true and openWorldHint=true; the description earns its place by disclosing that mutations are gated behind the AI Brewer's confirmation question, that identity verification happens remotely, and that the agent must relay the reply verbatim without adding promises, dates or refunds. That is exactly the non-obvious behavioural context the hints cannot express.

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 purpose, then identity, then conversation threading, then output-handling rules — a sensible order of operations with no marketing filler. Dense but every sentence imposes a rule the agent must follow.

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?

Only two parameters, no output schema, and the description still covers the full call lifecycle (first vs subsequent calls, identity prerequisites, confirmation gate, reply-relay constraint). Nothing an agent needs in order to invoke it correctly is absent.

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 0%, so the description must carry the load, and it does: 'message' is the customer's own words, and conversationId has a non-obvious lifecycle (omit on first call, echo thereafter). It misses the practical constraint that message caps at 1500 characters, which matters when pasting long customer text.

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 ('Help an IMPOSSIBREW customer with THEIR OWN orders and subscription') and enumerates the concrete task families: WISMO, tracking, one-time box, skip/pause/cancel, cancelling an unshipped order, reporting a problem. The emphasis on 'THEIR OWN' scopes it away from the product/catalogue siblings at a glance.

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?

Very clear about operational context: pass the customer's words, identity is checked downstream via email+postcode or order number (starts EC), you may look up their order emails but never guess, omit conversationId on first call and pass it back afterwards, and only act after the customer confirms. The one gap is that it never names when to route to ask_impossibrew instead, despite that being the obvious neighbouring assistant.

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 traits (readOnly, idempotent, openWorld, non-destructive). The description adds meaningful behavior beyond annotations by clarifying that it returns product resource links and that live facts come from resource readback, not from this call directly.

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 two tight sentences with no wasted language. It front-loads the typeahead purpose and follows with the key readback caveat.

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 low-complexity, single-parameter tool with no output schema, the description provides enough context: it returns product resource links rather than full product data, and live facts require readback. Minor gaps remain around result cardinality and exact query semantics.

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 0%, so the description must compensate. It partially does by indicating that the query matches IMPOSSIBREW beer names, but it does not explain matching behavior, formatting, or the 120-character limit that appears in 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 names a specific behavior: composer typeahead returning product resource links for matching IMPOSSIBREW beer names. It clearly distinguishes this from retrieving full product details via readback, though it does not explicitly contrast with siblings like browse_beers or get_product.

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?

"Composer typeahead" implies the context in which this tool is useful, but the description gives no explicit when-to-use or when-not-to-use guidance relative to browse_beers, get_product, or compare_beers. The intended usage is inferable but not fully spelled out.

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