Hydrogen Studies

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

Annotations already cover the safety profile (readOnly, non-destructive, closed-world). The description adds a genuine behavioral fact beyond them: unknown values are counted separately, which materially affects how an agent interprets bucket results. It says nothing about bucket ordering or the default top limit, but the schema carries those.

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 sentence, front-loaded with the core action and grouping facets, then the use case. 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 no output schema, the description does reasonably well by implying a bucket-to-count result shape and explaining unknown-value handling. Minor gaps remain around what a bucket looks like and default limiting, both of which are partly covered elsewhere.

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 73%, so most parameters are self-documenting. The description enumerates the group_by values (already in the enum) and adds a cross-reference to list_conditions for the condition slug, but several filters such as min_sample_size and the year range receive no extra 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?

States a specific verb (count) and resource (studies) plus the eight available grouping facets, and frames the intent as "trends and 'how many' questions". This clearly separates it from retrieval-style siblings like search_studies and get_study.

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 a clear use context (trends, count questions) and points to list_conditions for resolving condition values. It does not name the alternative tools or state when not to use aggregation, so no explicit exclusion.

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 this is a read-only, non-destructive, closed-world tool, so the safety profile is covered. The description adds a genuinely useful piece of behavior in the merge rule ('each group is a set of filters merged over the optional shared filters'), but it never says what an 'evidence profile' actually returns, and there is no output schema to fill that 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?

Two compact sentences, with the core operation front-loaded and the examples immediately after. There is no filler, but the second sentence is dense enough that it is easy to miss the merge rule on a skim.

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 nested 3-parameter comparison tool with no output schema, the description explains how input is composed but not what comes back per group: no mention of counts, pooled effect direction, or the top studies list. What it does say is accurate, but an agent still has to guess at the deliverable.

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?

With 67% schema coverage the schema already documents most filters, but the description adds semantics the schema does not: the 2–4 cardinality and the precedence relationship between per-group filters and the shared block. The optional top_studies parameter and its default are left entirely to the schema, which is a minor gap.

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 operation and resource: 'Side-by-side evidence profiles for 2–4 groups of studies'. Two concrete examples clarify what a comparison means here. It reads as distinct from the sibling aggregate_studies or get_condition_evidence, but it never explicitly says so.

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 examples ('inhalation vs drinking water for one condition, or two conditions') imply the comparison use case, but there is no explicit guidance on when to choose this over aggregate_studies or get_condition_evidence, and no stated 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 declare readOnlyHint=true, destructiveHint=false, and openWorldHint=false, so the safety profile is covered. The description adds the useful disclosure that results come from three distinct traversal directions, but says nothing about result limits, ordering, or how sparse the citation data may be.

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 efficient sentence that enumerates the three result categories with no filler. It leads with the content of the result set rather than a verb, which is slightly less front-loaded than an ideal framing, 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?

For a one-parameter read-only tool with full annotation coverage and no output schema, the description does the important work of explaining what the returned relations are. Missing only edge-case behavior (unknown identifier, empty citations), which is minor here.

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 100% and the 'study' parameter already documents that it accepts an id, slug, DOI, or URL. The description only refers vaguely to 'a given study' and adds no syntax, format preference, or ambiguity handling 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?

The description spells out exactly what the tool returns: cited studies, citing studies, and similar studies by shared health conditions. That three-way scope clearly separates it from get_study (a single record) and search_studies, 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 only implied — an agent can infer this is the tool to reach for when it wants literature connected to a known study. There is no explicit 'use this when…' statement and no guidance on when to prefer search_studies or get_condition_evidence instead.

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=true, destructiveHint=false and openWorldHint=false, so safety is covered. The description adds real context beyond that by spelling out the aggregation output shape (subject-type split, RCT/participant counts, publication years, data coverage, strongest studies), which is valuable given there is no output schema.

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 front-loaded sentence with the purpose stated first, followed by a dense but purposeful enumeration of returned metrics. No filler, though the catalog-style list of output facets is long enough to approach the edge of readability.

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 3-parameter aggregate with no output schema, the description covers what an agent needs to know about the response content. It stops short of covering behavior for an unknown/ambiguous condition or how large the returned payload is, but nothing critical 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?

Schema coverage is 100%, so the baseline is 3; the schema already documents condition, subject enum, and top_studies with defaults and bounds. The description hints at the subject filter ('human vs animal vs in-vitro') and top_studies ('strongest studies') but adds no syntax or behavior beyond those field labels.

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-and-resource framing: it produces an 'evidence profile' for a named health condition hub, then enumerates the profile contents (study counts, subject mix, RCT/participant counts, outcomes, delivery, years, coverage, top studies). This clearly separates it from get_study and search_studies, though it never names a sibling 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 only implied: you call it when you want an aggregate profile for a condition. There is no statement of when to prefer compare_evidence or aggregate_studies, nor any prerequisite note in the description itself (the 'see list_conditions' hint lives in the schema, not the description).

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 readOnlyHint=true, destructiveHint=false and openWorldHint=false, so safety is not the description's burden. What the description does add is the shape of the payload (counts, breakdowns, ranges, coverage), which matters because no output schema exists. It says nothing about caching, snapshot freshness, or how missing/unknown values are represented, but the core disclosure is present.

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 sentence fragment with no filler, front-loading the highest-level characterization ('size and makeup') before the itemized contents. Nothing could be cut 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 zero-param, read-only summary tool with no output schema, the description covers the main return buckets well enough to decide whether the result is useful. Terms like 'field coverage' and 'makeup' remain undefined, so the agent still cannot predict exact field names or value formats.

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?

Zero parameters, so there is nothing for the description to disambiguate; the schema is empty and coverage is moot. Baseline 4 applies for a parameterless tool.

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 enumerates the concrete contents of the resource — total studies, breakdown by subject and design, year range, field coverage — which is far more informative than the tautological title 'Database stats'. It lacks an explicit verb ('returns') and does not distinguish itself from the overlapping sibling aggregate_studies, which could also produce breakdowns.

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 when-to-use guidance, no prerequisites, and no named alternative. Given that aggregate_studies sits right next to it and plausibly produces similar breakdowns, the absence of any routing hint leaves the agent to guess which tool to call for corpus-level numbers.

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=true, destructiveHint=false, and openWorldHint=false, so the safety profile is covered; the description adds value by disclosing the shape of the payload (citation, H2 protocol, quality score with red flags, conditions/mechanisms, links). It stops short of describing lookup-failure behavior or whether multiple identifiers can be resolved ambiguously.

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, zero preamble, with the record's contents front-loaded ahead of the identifier formats. The long field enumeration is dense but each item is substantive information an agent would otherwise lack.

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 burden of explaining return contents and does so thoroughly across citation, protocol, outcome, critique, and score fields. The remaining gap is minor: nothing says what happens when an identifier does not resolve.

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 100% and the single parameter already documents the accepted id, slug, DOI, and URL forms. The description's "Accepts an id, slug, DOI or study URL" largely restates the schema rather than adding precedence or normalization rules, so the 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?

The description opens with a specific verb+resource ("Full record for one study") and enumerates exactly what the record contains, so an agent can tell this is a detail-fetch rather than a search tool. It never names a sibling such as search_studies or compare_evidence, so the differentiation is inferred from the word "one" 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 only implied: the agent must infer that this is the tool to call when it already has a study identifier and wants the complete record. There is no explicit when-to-use/when-not guidance and no statement of what to call instead when only a query is available.

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 (readOnlyHint=true, destructiveHint=false, openWorldHint=false), so the description's added value is the return contract: full enumeration ('Every'), per-entry study count, a nullable evidence grade, and a URL. The '(where reviewed)' qualifier usefully warns that evidence grade can be absent.

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 with zero waste — the return shape comes first, the consumption hint second. Nothing restates 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 of describing the response and does so by listing the per-entry fields and noting that the list is exhaustive. Nothing an agent needs in order to call it or interpret its result 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 tool takes zero parameters, so the schema baseline is 4 and there is no input semantics to explain. The mention of a `condition` filter refers to other tools, not this one, but it is correctly framed as guidance rather than an input.

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 resource (health condition hubs) and enumerates exactly what each entry carries: study count, evidence grade with a 'where reviewed' caveat, and page URL. It is clearly distinguishable from siblings like get_condition_evidence or aggregate_studies, though it names none of them 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 second sentence gives actionable downstream guidance — the returned slug is what you pass to the `condition` filter on other tools — which implies this is the discovery/enumeration step before a targeted query. It does not explicitly name which sibling to call next or when to skip this tool.

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=true and openWorldHint=false, so the safety profile is covered. The description adds meaningful context beyond that: the exact fields each compact record contains and the default ordering rule (relevance with a query, evidence strength without one). It does not mention pagination limits or result caps, which a 12-param search tool would benefit from.

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 with no filler; the retrieval purpose is front-loaded and the return-shape and sort behavior follow in a logical order.

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 12-param filter tool with no output schema, describing the returned record fields is exactly the right compensation. Remaining gaps are pagination behavior and filter-combination logic, which are secondary but would help an agent avoid misuse.

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%, so the schema carries most parameter meaning. The description adds real value for the undocumented `sort` parameter by explaining the default ordering behavior, and it echoes the condition-hub convention already in the schema. It says nothing about offset/limit pagination or year range semantics, leaving those to 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 ('Find molecular hydrogen studies') plus the search mechanism (keywords and/or filters) and the return payload. It does not distinguish itself from siblings like aggregate_studies, compare_evidence, or find_related_studies, so an agent must infer the split from tool names 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?

'By keywords and/or filters' implies a flexible retrieval use case, and the sorting sentence hints at query-vs-no-query modes. However, it never says when to choose this over sibling retrieval tools such as find_related_studies or get_condition_evidence, nor whether filters combine as AND or OR.

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