Cerulean Jobs

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

Annotations already declare readOnly, idempotent, non-destructive, non-open-world behavior, so the safety profile is covered. The description adds genuine value beyond them by disclosing the error case ('An unknown or closed listing is an error') and the return shape ('that one listing in full as text, with its url and structured details'). It stops short of describing pagination/size limits, but for a single-item fetch that is minor.

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-loads the when-to-use condition and the id prerequisite in the first sentence, then adds return and error behavior in order of importance. Slightly dense with the field enumeration, but every clause carries selection-relevant information and there is 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 only two well-documented parameters, full annotations, and an output schema covering return values, the description supplies exactly the missing pieces: trigger conditions, the no-search exclusion, and error semantics. Nothing an agent needs in order to call 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 100%, and the locale parameter is documented in unusual depth in the schema itself (BCP 47 format, site mapping, default). The description adds no parameter-level detail beyond the schema, so the baseline 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?

States a specific verb and resource ('fetch' one Cerulean listing) and enumerates the fields returned (responsibilities, qualifications, skills, pay, benefits, languages), so an agent can tell exactly what this returns. It also explicitly carves out what it is not: 'It does not search.'

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 an explicit trigger condition ('when someone asks about one specific listing') plus the prerequisite ('its id or Cerulean URL is known') and an exclusion ('It does not search'), which implicitly routes search-style requests to the search sibling. Nothing about when to choose this 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 readOnly, idempotent, non-destructive and closed-world, so the safety profile is covered. The description contributes genuinely new behavior: unknown listings or missing similarity data produce an error rather than unrelated fallback results, which tells the agent not to treat an empty-looking outcome as a soft failure.

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, zero filler, with the trigger condition and the id-not-keywords constraint front-loaded before the failure-mode note. Every sentence carries distinct 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 read-only similarity lookup with a full output schema, complete annotations and 100%-documented parameters, the description covers the remaining gaps an agent needs: when to reach for it, what input form it expects, and how it fails. Nothing material is left to inference.

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 already 100%, including a detailed `locale` and paging description, so the baseline is 3. The description adds only one piece of parameter meaning beyond the schema: that the input is a listing id rather than keyword text. That is useful but marginal.

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 (find open roles similar to one specific Cerulean listing) and scopes it by 'what the job is and where it is'. It also distinguishes itself from keyword search with 'Takes that listing's id, not keywords', so an agent can separate it from the `search` sibling without opening either 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?

'Use this when someone wants more roles like one specific Cerulean listing' gives a clear trigger condition, and 'not keywords' implicitly routes keyword-driven discovery elsewhere. It stops short of naming `search` as the explicit alternative, so the routing is inferable rather than spelled out.

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, non-destructive behavior, so the description is free to add the context an agent actually lacks: data provenance (houses' own career sites, re-read every four hours), that closed roles drop out, that results are one page at a time, and that a total match count is returned. These are non-obvious traits that change how an agent should phrase results and paginate.

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 when-to-use clause, then provenance, then parameter behavior, then return shape — a sensible order. It is a single dense paragraph for a 12-parameter tool, so the length is largely justified, though the parameter clauses echo the schema and could be trimmed.

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 zero-required-parameter, 12-parameter search tool with four enums, this covers the things an agent needs: freshness, closed-role exclusion, filter composition, and the shape of a result page (id, summary, apply url) plus total. With an output schema present it need not document return fields further, and nothing material is left ambiguous.

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 explains query interpretation, filter combination, clamping, and locale fallback in detail. The description restates the same facts (query accepts keywords or a plain sentence, filters combine, posted_within_days narrows recency) rather than adding syntax or precedence rules beyond the schema. 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 set (find, browse, count, compare) against a specific resource (open roles at luxury houses and brands) and bounds the corpus (career-site listings, refreshed every four hours, closed roles removed). An agent can distinguish this from find_similar_jobs or fetch by the breadth of the search surface, but the description never names those siblings 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?

"Use this when someone wants to find, browse, count or compare open roles" gives clear triggering context, and "query may be empty to browse" plus "filters combine" tells the agent how to drive the tool. Missing is any when-not guidance or a pointer to find_similar_jobs for similarity-style requests, so the routing decision against siblings 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.