SOTA2 — AI Solution Rankings

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

Annotations already cover readOnly/idempotent/openWorld safety, but the description adds real context: dimensions default to all, each collection paginates independently with the same limit/offset, missing catalog records may still carry ranking or revenue data, and soc2Disclosure values are vendor claims not certifications. That is meaningful behavioral disclosure beyond the 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?

Front-loads the core verb and resource, then layers dimension and pagination details efficiently in five short sentences. The soc2Disclosure sentence is dense and slightly cryptic but carries non-redundant meaning.

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?

An output schema exists so return shape needn't be described; the description covers dimension selection, default behavior, pagination, and a subtle data-integrity caveat. Complete enough for correct invocation, with only sibling routing left implicit.

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 75% and the schema has no description for the dimensions array, so the description earns credit by enumerating the available dimensions and stating the default-to-all behavior. It also restates the shared limit/offset pagination model, though it adds little beyond the schema's own limit/offset descriptions.

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 (Returns) and resource (public data for an exact company ID or slug), and the word 'exact' implicitly distinguishes it from the search_* siblings. It does not name the sibling it replaces, so it stops 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?

'Exact company ID or slug' implies you must already know the identifier and therefore should resolve names elsewhere, but no alternative is named and no explicit when-not guidance is given. Usage is inferable rather than stated.

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 and open-world behavior, so the lower bar applies. The description adds real semantic context beyond that: what the response contains (ranks, ties, USD revenue, confidence, sources) and the warning that revenue, ARR, run-rate and forecasts carry distinct basis labels, which prevents an agent from conflating metrics.

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 purpose, then payload, then the caveat. No filler; the example IDs and basis-label warning each earn their place.

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?

An output schema exists, so return value detail is optional and the description's summary is a bonus. The only real gap is not pointing an agent at list_company_leaderboards to obtain valid IDs, which matters for a required exact-match 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 coverage is 67% and the required 'leaderboard' parameter has no description there, so the description meaningfully compensates by giving concrete example IDs (ai-global, ai-usa, ai-china). The limit/offset pagination semantics are already well documented 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?

States a specific verb and resource ('returns a published AI company leaderboard') plus the key scoping constraint that it takes an exact ID. It doesn't explicitly name how it differs from list_company_leaderboards or get_solution_rankings, so an 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?

'By its exact ID' implies this is a lookup rather than a discovery tool, but the description never says to call list_company_leaderboards first to find IDs, nor when to prefer get_solution_rankings. Usage is only implied.

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 (readOnly/idempotent/openWorld), disclosing that filtering is applied to the complete ranked benchmark before pagination so positions stay global, that ties are preserved, and that rows are never deduplicated into method families. It also flags that responses include metric direction, completeness checks, invalid-percentage warnings and separate reporting/origin papers, which is exactly the behavioral context an agent needs before calling.

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?

Dense and front-loaded — the core selection semantics come first, followed by filtering/pagination behavior, then response contents. A couple of clauses ('Top N corresponds to limit N rows') duplicate the schema, but nothing is padding.

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 an output schema present and full annotation coverage, the description needn't explain return shapes, and it still covers pagination/continuation, global-position stability under filtering, tie handling and non-deduplication. An agent can call this correctly without further 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%, and the description still adds cross-parameter meaning: limit is framed as 'Top N' rows and also as the page size for continuation, and the method substring's interaction with offset is restated so positions are understood as global. That is genuine added value, though it largely restates the schema's own wording rather than introducing new semantics.

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 concrete verb and resource: it returns individual reported result rows for one exact benchmark version/split and metric, with each row carrying resultId, configuration and global position. The clause 'not deduplicated or grouped into method families' implicitly contrasts it with an aggregation/ranking sibling, but no sibling is named explicitly, so differentiation is inferred rather than stated.

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 prescribed: the description explains that continuation should use results.nextCall and that a different metric starts at offset 0, and the schema points to discovery for the benchmark slug. It never says when to choose this over get_solution_rankings or get_company_leaderboard, nor when not to use 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?

Beyond the readOnly/idempotent/non-destructive annotations, the description discloses error semantics an agent needs: missing entities return not_found rather than a tool error, while API failures do raise errors. It also previews what each entity type returns (benchmark links, abstract/authors/PDF/code links, related benchmarks).

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?

Tight, front-loaded block with no filler sentences; each clause carries distinct information (identifier formats, per-type payloads, error behavior). It is dense to the point of being a single wall of text with four topics packed in, which slightly hurts scanability.

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 an output schema present, the description needn't restate return shapes, yet it adds useful per-type summaries and distinguishes not_found from tool errors. An agent has everything needed to call it correctly; only the pagination parameter remains unaddressed.

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 only 33%, but the description compensates well for the two required parameters: dataset identifiers must be family slugs not version slugs, papers use 24-character IDs not arXiv IDs, and task/domain use canonical slugs. Only the optional 'page' parameter is left 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?

Opening sentence names a specific verb and resource (returns one exact research entity) and enumerates the four entity types. The word 'exact' signals it is a single-item lookup, which implicitly separates it from search_research, but that sibling is never named.

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: 'one exact research entity' plus the identifier-format rules tell an agent this is a direct lookup requiring a canonical ID, so discovery-type queries belong elsewhere. No explicit when-not-to-use or named alternative (search_research, list_tasks) is given.

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/idempotent/openWorld annotations by disclosing failure modes: missing entities return not_found with discovery guidance, ambiguous categories return needs_selection or clarification_required, and results never silently fall back to another scope. It also warns that positions/scores apply only within the selected scope and that cross-scope quality comparisons are not established.

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 core purpose in the first sentence and packs the scope matrix into the next few, with no filler. Slightly telegraphic and one clause ('Top N corresponds to limit N rows') largely repeats the schema's limit description.

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 an output schema present it needn't describe return values, and the annotations cover the safety profile, so what remains is scope and error semantics — both covered well. The one gap is the zero-required-parameter case: the description never says what happens when neither industry nor task is supplied.

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 50% (limit and offset are documented; task and industry are bare strings), and the description compensates by explaining what each scope means and how the two combine. It also confirms limit maps to top-N rows, reinforcing the schema's limit text. It does not explain the behavior when both task and industry are omitted.

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 published AI solution rankings' scoped by industry/task. The resource (AI solution rankings) is clearly distinct from sibling tools like get_company_leaderboard and get_research_benchmark, though it never names those siblings to make the contrast 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?

Gives explicit scope-selection rules: industry alone returns the task directory and cross-industry links, industry + task selects the exact scope, task alone selects cross-industry. This is strong guidance on how to combine parameters, but it doesn't say when to prefer this tool over get_research_benchmark or get_company_leaderboard.

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. The description still earns credit by disclosing the returned shape (IDs, scope, descriptions, source URLs) and, more importantly, a non-obvious semantic caveat that country lists rank companies rather than countries — a point no structured field carries.

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 with no waste, leading with what is listed before the return shape and the caveat. Each sentence contributes something the agent would otherwise have to guess.

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 an output schema present and zero parameters, the description need not explain return structure or arguments, and it does not overreach. The one gap is that it never connects this discovery call to get_company_leaderboard, which would complete the workflow for an agent.

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, which sets the baseline at 4; there is nothing for the description to disambiguate. The stated output fields are extra context rather than a substitute for parameter documentation.

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 ('List supported AI company leaderboards') and names the dimensions available (revenue, country/region, category). The list-vs-get split against the sibling get_company_leaderboard is implied by 'supported' but the sibling is never named, so differentiation is inferential rather than 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 only implied: an agent can infer this is the discovery call that yields IDs for get_company_leaderboard, but that routing is never stated. There are no explicit when/when-not conditions or named 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 already declare readOnly, idempotent, non-destructive, open-world, so safety is covered. The description adds genuine behavioral context beyond that: pagination as a browse mode and the conditional presence of cross-industry/consumer/agent audience data ('when available'), which tells the agent not every industry will carry every field.

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 sentences, each load-bearing: output shape first, conditional data second, filtering/pagination mode last. No restatement of the title and 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?

For a read-only, zero-required-param listing with an output schema and strong annotations, the description covers what to expect and how to page. The only material gap is the query's matching semantics, which neither schema nor description resolves.

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 67%: limit and offset are documented in the schema, while query has only min/max length constraints. The description partially compensates by calling query an optional directory filter, but it never says what the query matches on (name, slug, alias?), which is the one thing an agent most needs to formulate a valid query.

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 ('lists supported industries') and enumerates the returned fields (IDs/slugs, names, descriptions, aliases), so the agent knows exactly what this produces. It does not explicitly contrast with any sibling, but no sibling is a directory lister, so the distinction is clear from the description alone.

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 last sentence gives clear operational context: use an optional query to filter, or omit it and browse with pagination. That is real when-to-use guidance, though it never names an alternative tool or a when-not-to-use condition.

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 open-world behavior, so the safety profile is covered. The description adds that the result is a directory of canonical slugs plus descriptions and that filtering is optional, but says nothing about result size, ordering, or whether the full list is returned when no query 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?

Three tight sentences with the core purpose first, then the disambiguation from industries, then the filter behavior. No filler, though the slug/description enumeration is slightly redundant with the examples.

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?

Output schema exists, so return shape need not be explained, and the tool is a simple single-parameter listing. What is missing is filter-matching behavior and result-set expectations, minor gaps for a discovery 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 coverage is 0%, so the description has to carry the single parameter, and it does state that 'query' filters the domain directory. It does not describe matching semantics (substring vs. exact vs. fuzzy) or what the query matches against, so it only partially compensates.

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 ('Lists AI research fields with canonical slugs and descriptions') and gives concrete examples of the domains returned. It explicitly distinguishes these research domains from commercial company industries, which separates it from the sibling list_industries.

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?

Clear that the domain directory can be filtered with an optional query, and the industries exclusion routes the agent away from list_industries. It does not say when to prefer this over search_research, but the browse-vs-search split is reasonably implied.

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, so the bar is lower, yet the description still contributes: it discloses a clarification_required response branch for ambiguous queries and warns that company coverage overlaps so task counts are not additive. It omits anything about result volume or continuation behavior beyond what the schema states.

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 sentences, front-loaded with what is returned before the conditions. Dense but each sentence carries distinct information (scoping, ambiguity behavior, non-additive counts); 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?

An output schema exists, so return values needn't be re-explained, and the description covers the two non-obvious behaviors (clarification_required, overlapping coverage). Remaining gap is how query/industry values are expected to be formed, which matters for a 4-param open-world lookup.

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%: limit and offset are documented in-schema, while query and industry ship bare. The description compensates partially by explaining query is a free-text directory filter with ambiguity handling and industry is a scoping filter, but it gives no syntax for either (name vs slug, matching rules).

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 plus the returned fields: 'Lists AI solution task IDs, slugs, descriptions and company coverage'. It is clearly not one of the company/research siblings, but it never explicitly contrasts itself with list_industries or search_research, which the 'industry' and 'query' params make adjacent.

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: use it to enumerate tasks, optionally scoped by industry, with a free-text query filter. It adds real guidance about ambiguous queries ('verification', 'verify') triggering clarification_required, but never says when to pick this over list_industries or search_research, nor any preconditions.

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, non-destructive, open-world, and the description adds real behavior beyond them: revenue-only records with null catalog slug and profile URL whose ID is still usable, alias/spelling-dependent matching, ambiguous names yielding multiple matches, and a clarification_required outcome for verification-style queries. These are exactly the failure modes an agent needs before calling.

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 is front-loaded and most sentences carry distinct, non-redundant information. The final sentence about generic verification queries is densely compressed and slightly hard to parse, and pagination/continuation is never surfaced despite being mentioned in the limit 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?

An output schema exists so return format needn't be spelled out, yet the description still flags the nullable-slug/profile-URL case and the clarification_required response — useful context. What's missing is how to resume a paged search and any hint about result ordering or ranking.

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%: limit is fully documented in-schema while query has only minLength/maxLength constraints. The description adds meaningful query semantics (matching depends on spelling and available aliases, ambiguous names produce multiple matches) but never states what a valid query should contain or how to continue paging, which the description leaves to the limit field's schema 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: 'Resolves a named AI company to stable IDs, names and descriptions.' It also explicitly separates itself from category-leaderboard siblings ('searches company identities rather than category leaderboards'), so an agent can route without opening schemas. Multiple-match behavior on ambiguous names confirms it is a search, not a single-record fetch.

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 clear use conditions and one explicit exclusion (identity search vs. leaderboards), plus a when-not case: generic verification queries return no rows and trigger clarification_required. It does not name specific siblings like get_company or get_company_leaderboard, though 'returned ID remains usable for company lookup' implies the handoff.

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 the safety profile (readOnly, idempotent, openWorld, non-destructive), yet the description adds real behavioral context: 20 results per API page, that versions/test splits/Full-Hard variants may appear as separate benchmarks, and that search summaries are discovery data rather than validated ranks. That last point meaningfully warns the agent against over-reading results, which annotations cannot convey.

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?

Dense but economical — the resource scope and match keys are front-loaded, then constraints follow. The two closing sentences about discovery data vs performance ranks overlap somewhat but address distinct pitfalls (summary semantics vs sort ordering), so little 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?

An output schema exists, so return values need not be explained, and the annotations carry the safety profile. Still, for a 7-parameter search tool with no schema descriptions, the description should explain what domain/task/dataset filters actually do and how resource interacts with query; that gap keeps it at minimum-viable completeness.

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 must compensate, and it partially does: it discloses page size (20), the query requirement for paper search, and the semantics of count-based sorts. However, domain, task, dataset, resource (default 'benchmarks'), and the page bound of 1–1000 are left entirely undocumented, so several parameters remain opaque.

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 (searches) over a clearly enumerated set of resources (benchmarks, datasets, tasks, papers) and names the match keys (keywords, names, titles, authors, arXiv IDs). This cleanly distinguishes it from sibling detail-getters like get_research_benchmark and get_research_entity, which retrieve a single entity rather than discover across resource types.

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?

Offers one concrete conditional rule — 'Paper search requires query and has no count/date sort' — which tells the agent a prerequisite for a resource type. But it never states when to use this discovery tool versus the sibling retrievers (get_research_entity, get_research_benchmark) or when to switch tools after a search. Usage is implied rather than prescribed.

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