Swervie

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

Beyond the readOnlyHint annotation, the description discloses important behavior: duplicate seller listings are collapsed, hard filters exclude rows missing the attribute, soft preferences only re-rank without excluding rows, default sort is value, and price-dimension filters apply to live offer price in cents. It also notes that buy_url responses carry a disclosure field, which is substantive behavioral context not covered by annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is front-loaded with the core purpose and then layers in row contents, filter semantics, sort options, price filtering, and disclosure behavior. It is dense but mostly earns its length; the enumeration of row fields is slightly long but still useful for understanding the return shape.

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 complex 7-parameter tool with no output schema, the description covers the return shape, deduplication, hard versus soft filtering, sort behavior, and commission disclosure. It leaves pagination details such as cursor behavior and the limit maximum/default unspecified, which are minor gaps but relevant for correct invocation.

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 low at 29%, but the description compensates substantially by explaining sort options, hard versus soft filter semantics, Price-dimension min/max behavior, and preference examples such as 'Rated Noise, prefer low' and a pet-hair rating, prefer high. It does not explain cursor or limit behavior, leaving some parameter semantics undocumented.

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: returns a category's consolidated, ranked shelf with duplicate seller listings collapsed. This clearly distinguishes it from sibling tools like search_products or find_category by describing the output as a deduplicated, ranked category shelf rather than a search or lookup.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

Usage is implied by 'Return a category's consolidated, ranked shelf' and by explaining hard filters versus soft preferences, but it never explicitly says when to use this tool instead of search_products, find_category, or get_products. The sort and filter semantics help an agent use the tool once selected, but tool selection guidance 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 declare readOnlyHint and openWorldHint, so the safety profile is covered; the description still adds real value by disclosing the exact shape of the payload (filter specs with type/units/allowed values/importance plus price bands), which is essential since there is no output schema. It stays silent on lookup failure behavior or unknown-intent handling.

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 action and then enumerates the return contract. Every clause earns its place, though the enumeration of filters is slightly list-heavy.

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 dutifully describes the returned contract, which is the main completeness risk for this tool. The remaining gap is sibling disambiguation and the semantics of an unmatched intent.

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%: user_signature is thoroughly documented in the schema, while query is bare. The description only indirectly characterizes query as a shopping intent string, adding modest meaning beyond the schema and not compensating for the undocumented 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?

Clear verb ('Find') plus a specific resource ('Swervie product category') and an explicit statement of what is returned ('its filter contract'). It does not differentiate itself from the sibling browse_category, so the agent must infer the boundary.

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 phrase 'matches a shopping intent' implies usage (natural-language intent in, category contract out), but there is no explicit when-to-use versus browse_category/get_products/search_products and no stated prerequisites 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 only declare readOnlyHint and openWorldHint; the description adds substantive behavior beyond that — the 25-item cap, spec ordering (card_specs first), group-row id substitution with resolved_groups reporting each substitution, and a commission disclosure field accompanying buy_url links. It stops short of stating what happens with unknown or partially-resolvable ids, which is the one behavior an agent would still have to guess.

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 the headline ('Full detail for up to 25 products') followed by a dense enumeration of returned fields. It is long but nearly every clause adds information; only the disclosure sentence could be tightened without loss.

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 carries the full burden of describing the response and does so thoroughly: spec ordering, ratings with evidence quotes, all offers with price and buy_url, attribute_provenance, resolved_groups, and the disclosure field. For a read-only, two-parameter tool this is enough for an agent to call it and interpret the result.

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% — user_signature is fully documented in-schema while ids is a bare string array. The description compensates by explaining that ids accepts both product ids and shelf row ids and how group-row ids resolve to top_pick_id, which is genuine meaning beyond the schema's maxItems: 25.

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 with scope: 'Full detail for up to 25 products', then enumerates exactly what detail is returned (specs, ratings, offers, provenance). It contrasts implicitly with summary-style siblings like search_products, but never names lookup_product or search_products to make the boundary 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?

Usage is implied rather than stated: an agent infers this is the detail-expansion step after obtaining ids from search or browse. There is no explicit 'use this instead of lookup_product when...' or any guidance on when to prefer the single-product sibling over this batch call.

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 readOnlyHint/openWorldHint annotations, the description discloses the match outcome taxonomy (exact/likely/ambiguous/none), what runner_up and rows contain, that misses are recorded as demand, that unavailable-price products are still returned under a readiness flag, and that buy_url responses carry a commission disclosure. That is substantive behavior an agent could not infer from structured fields.

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?

One dense but front-loaded paragraph that moves from inputs to match semantics to return contents to edge cases, with no filler sentences. It is on the long side, though every clause carries information the 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?

With no output schema, the description carries the full return-value burden and does so: it explains match values, products[0] contents, shelf_context, readiness, and the disclosure field. Nothing essential for correct invocation or interpretation appears 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?

Schema description coverage is 80%, so the schema already documents query, category, identifiers, and user_signature. The description adds marginal detail — that a barcode may be embedded in the name string and that brand+model resolves mpn/model_number — but says nothing about user_signature's quota/attribution role or category scoping beyond what the schema states. Baseline 3 fits.

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 ('look up one specific product') and enumerates the accepted identifier forms (name, ASIN, UPC/EAN/GTIN, brand+model), which makes the shape of the operation unambiguous. It does not explicitly contrast itself with search_products/get_products, so sibling differentiation is left to inference.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

The enumerations of acceptable inputs imply when this tool applies (you know the exact product or an identifier). But there is no stated 'use search_products instead when you only have a vague need' exclusion, and no mention of prerequisites such as needing at least one of query/brand/identifiers despite zero required parameters.

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 beyond the readOnly/openWorld annotations: it defines match_quality tiers, states which tiers return no rows, describes demand queuing (queued_for_creation, queued_product_request), and discloses the commission disclosure field on buy_url responses. That is exactly the kind of side-effect and side-channel context an agent needs.

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-loading is good – the verb and concrete examples come first – but the remaining text is a dense, unbroken block of response-shape and mode vocabulary. Most sentences earn their place, yet the lack of structure makes it harder to scan than its length alone would justify.

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 carries the burden of explaining the return shape (products[0], shelf_context, interpretation, card_specs, disclosure), which it does in detail. The omission of pagination and sorting behavior is the main remaining gap for a tool this complex.

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 only 20%, so the description must compensate, and it does thoroughly for the query parameter's semantics (mode resolution, hard vs soft constraints). However, sort, limit, and cursor are never mentioned in either schema or description, leaving three parameters unexplained.

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 ('Answer one free-text shopping query') and clarifies the query-shape modes (product, comparison, category). It never names siblings like lookup_product or browse_category, so the agent must infer the split itself, which keeps it 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?

Four example queries plus explicit mode triggers ('A query naming one specific product...', "'A vs B'...", 'Otherwise the category is resolved') give clear context for when each behavior applies. It offers no explicit exclusions or pointer to an alternative tool for exact-SKU or browse use cases.

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