Wardrobe Connect

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

Annotations already cover the safety profile (readOnly, idempotent, openWorld, non-destructive), but the description adds substantial behavior beyond them: result ordering (closest match, then cheapest), match criteria (colour, fabric, garment kind, gender; kids only on request), and hard exclusions (adult/fetish/costume never shown, duplicates collapsed, suspiciously cheap listings filtered). This is exactly the extra context annotations cannot carry.

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 purpose and an example before moving to inputs and then return/filter behavior. It is dense but every clause carries real information; the "it needs nothing else from the conversation" restatement is the only slightly redundant phrase.

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: eBay listings in the Clothing/Shoes/Accessories category, ranking rules, plus search links for other shops and a Facebook Marketplace search phrase. Combined with the input coverage, 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% and each parameter (q, size, country, condition, max_price) is already well documented with formats and defaults. The prose merely enumerates the same optional parameters ("a size, a price limit, the condition and a country") without adding syntax, units, or edge-case behavior beyond the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

Specific verb (find) plus resource (clothes) with concrete example queries ("black pleated wide-leg trousers"), and it scopes the input to a single item's short description. That scoping implicitly separates it from the image-based sibling find_clothes_by_image.

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 clearly when to use it (a short text description of one item) and explicitly says it "needs nothing else from the conversation", which is useful routing guidance for an agent holding extra context. It never names find_clothes_by_image or states the text-vs-image selection rule explicitly, so it falls short of a 5.

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, openWorld, non-destructive), but the description adds real behavioral context beyond them: the result set is restricted to the Clothing/Shoes/Accessories category, ordering follows eBay's look-alike order, and the failure mode is disclosed (if the picture doesn't show one piece of clothing, `matches` is empty and the say line asks what the item is). It also explains that q changes ranking semantics. No mention of rate limits or image-fetch failures, which is the only remaining gap.

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 purpose and scope, and every sentence carries information (inputs, return scope, failure mode, ranking effect). The phrasing is conversational ('the say line', 'it needs nothing else from the conversation') and could be tightened, 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 six parameters, no output schema, and annotations covering only safety, the description does the heavy lifting: it describes the source and ordering of results, the category restriction, and the empty-result behavior keyed to a `matches` field. It only references the response envelope ('matches', 'say line') in the failure case rather than describing the normal return shape, which is the one incompleteness.

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, but the description adds meaning the schema does not: it explains that `q` is what steers results toward the right item and 'removes look-alikes of the wrong kind,' and that image_url must be a direct https link to a single picture. That is genuine semantic lift over the field docs.

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 (find clothes visually matching a picture) and pins down scope: 'visually similar eBay listings from the Clothing, Shoes and Accessories category only, in eBay's look-alike order.' The image-input framing makes it trivially distinguishable from the text-based sibling find_clothes, even without naming it.

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?

Clearly signals when to use it (you have a picture of one item) and what is optional versus required: 'Takes the picture's direct https link and, if wanted, a few words naming the item, a size, a price limit, the condition and a country; it needs nothing else from the conversation.' That last clause is useful guidance to not ask follow-up questions. It never states when NOT to use it or points to find_clothes as the text alternative.

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