The Aggregate — LLM benchmark aggregate

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

With no annotations, the description carries the full burden. It discloses the tool returns explanations of IRT fusion, coverage counts, update cadence, and citation info. For a read-only informational tool, this is sufficient, though it does not describe response format or potential restrictions.

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 a single concise sentence that front-loads the purpose ('What this data is') and lists four key areas. Every word earns its place; no unnecessary 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 an informational tool with no parameters and no output schema, the description covers the main content areas. It does not provide an example or specify the structure of the output, but it is sufficiently complete for the tool's simple nature.

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 has zero parameters, so no parameter information is needed. The baseline for no parameters is 4. The description does not mention parameters, which 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?

The description clearly states the tool provides information about the aggregate: how IRT fusion works, coverage counts, update cadence, and citation. This is specific and distinct from sibling tools like get_leaderboard or get_model, which focus on querying data rather than explaining 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?

The description implies use when background information about the aggregate is needed, but it does not explicitly state when to use this tool vs alternatives or provide context exclusions. Given the self-evident purpose, this is adequate but not exemplary.

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?

No annotations exist, so the description fully discloses outputs and mentions a significance calculation. It does not note authentication, rate limits, or side effects, but given the read-only nature of comparisons, it is adequate.

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 a single, front-loaded sentence that efficiently conveys purpose and outputs with zero wasted words.

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?

The description covers the main outputs for a comparison tool. Without an output schema, it sufficiently explains return values. It lacks error handling details, but the tool is simple with one 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?

The schema already fully describes the models parameter (100% coverage). The tool description repeats the min/max but adds no new semantic information beyond 'two to four model names or slugs'.

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 specifies a head-to-head comparison of 2-4 models, listing exact outputs (aggregate ranks, Elo gap with significance, notable benchmarks). It clearly distinguishes from sibling tools like get_leaderboard (overall rankings) and get_model (single model details).

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 description implies use when comparing specific models, but does not explicitly state when not to use it or mention alternatives. However, the context of sibling tools makes the purpose clear.

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?

With no annotations, the description carries full burden. It lists all returned fields (measure, URL, IRT stats, skill weights, top models), implying a safe read operation. It lacks explicit mention of side effects or auth, but for a read-only lookup, transparency is adequate.

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?

Single sentence listing all returned items is efficient and front-loaded. Slightly dense at 29 words, but no redundancy. Could be split for readability, but it's clear and compact.

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?

No output schema exists, so description must cover return values. It lists key items (measure, URL, IRT stats, skill weights, top models). Parameters are well-documented in schema. For a simple retrieval tool, this is sufficiently complete for an AI agent to understand the output.

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% with clear descriptions for both parameters ('benchmark' and 'top'). The tool description adds no extra parameter details beyond summarizing the output. Baseline 3 applies as schema does the heavy lifting.

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 clearly states the tool retrieves detailed information for a single benchmark, listing specific data points (measure, URL, IRT stats, skill weights, top models). This distinguishes it from sibling tools like get_leaderboard (list) or search_benchmarks (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?

Implicitly, the tool is for deep-diving into one benchmark, contrasting with sibling tools for listing or searching. However, no explicit guidance on when to use or avoid is provided, missing a clear when-not or alternatives statement.

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?

With no annotations, the description carries the behavioral disclosure burden. It explicitly states the default one-row-per-model behavior, the include_variants option to split by reasoning-effort settings, and paging via limit/offset. It also adds methodological context (Elo scale, IRT model) that sets expectations, though it stops short of declaring read-only status or data freshness.

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 two dense sentences that front-load the core purpose ('Top of the cross-benchmark aggregate ranking') and include only essential details: methodology, default row grouping, and pagination. Every clause earns its place with 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?

The description covers purpose, methodology, default behavior, and paging, which is sufficient for an agent to select and invoke the tool correctly. It lacks explicit return field names and error behavior, but given the simple schema and absence of an output schema, it is nearly complete for a leaderboard listing 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?

The schema covers all three parameters with 100% coverage, so the baseline is 3. The description adds value by linking include_variants to the default fusion behavior and explaining that paging uses limit/offset, complementing the schema without redundancy.

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 clearly defines the tool's role as returning the top of a cross-benchmark aggregate ranking, with a specific methodology (IRT model over ~5,000 benchmarks). This distinguishes it from sibling tools like get_benchmark (single benchmark) and get_model (single model) by focusing on the aggregate perspective.

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 description implies usage by describing the aggregate ranking and default row-per-model fusion, and mentions paging, but it does not explicitly contrast with sibling tools or state when not to use this tool. For example, it does not say 'for a single benchmark, use get_benchmark 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?

No annotations are provided, so the description must fully disclose behavioral traits. It describes the output content but does not mention idempotency, authentication needs, rate limits, or any side effects. As a read operation, it likely has minimal impact, but this is not confirmed.

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 a single, front-loaded sentence that efficiently lists the tool's output. It is concise with no redundant words, though it could be slightly more structured (e.g., bullet points).

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?

Despite lacking an output schema, the description enumerates the key return fields: rank, Elo, provider, cost, and benchmarks. For a single-parameter tool, this provides a clear picture of what the tool returns. Missing details like output format or pagination are not critical 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 description adds minimal value beyond the schema: the parameter 'model' is described in the schema as 'Model name or slug, e.g. "Claude Opus 4.5" or "gpt-5-5". The description does not add new semantic detail (e.g., case sensitivity, format constraints), so a baseline of 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?

The description clearly states the tool provides in-depth profile for one model, listing specific data: aggregate rank, Elo, provider, cost, and notable benchmarks. This distinguishes it from siblings like 'compare_models' (multiple models) and 'get_leaderboard' (list).

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 description implies the tool is for obtaining detailed single-model info, but it does not explicitly state when to use it over alternatives like 'compare_models' or 'search_models'. No when-not-to-use or prerequisite guidance 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?

The description explains the purpose and return value but does not explicitly disclose behavioral traits such as read-only nature or side effects. Since no annotations are provided, the description partially carries the transparency burden, but it is adequate for a no-parameter tool.

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 concise, consisting of two sentences. The first sentence provides context, and the second states the output. Every word earns its place with no wasted text.

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?

Given the tool has no parameters and no output schema, the description is sufficiently complete. It explains the concept and return value. Minor gaps exist, such as not clarifying 'wins and losses' detail, but it remains highly functional for a simple 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?

The input schema has zero parameters and 100% schema description coverage, so the description does not need to add parameter details. The baseline score of 3 applies, and no additional value is provided 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 clearly states that the tool returns the current monthly standings of a prediction duel, including wins and losses. It uses a specific verb and resource, distinguishing it from sibling tools like get_leaderboard or compare_models.

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?

No guidance is provided on when to use this tool versus alternatives. The description explains what it does but does not specify contexts or exclusions, leaving the agent to infer usage implicitly.

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?

With no annotations, the description must carry behavioral context. It states the tool returns three fields and searches 'by (partial) name', but does not disclose aggregation semantics, sort order, case sensitivity, or pagination behavior beyond the limit parameter.

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 a single, front-loaded sentence that efficiently conveys the action and returns. It is concise but could be slightly improved by structuring the return values as a bullet list.

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 simple search tool with 2 well-documented parameters and no output schema, the description is adequate. However, it lacks clarity on what 'in the aggregate' means and does not specify matching behavior (e.g., case sensitivity, fuzzy vs exact).

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?

Since schema coverage is 100% and both parameters have clear descriptions, the tool description adds no additional meaning beyond stating 'by (partial) name', which repeats the query parameter description. 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?

The description clearly states the tool finds benchmarks by partial name and lists the return fields (model coverage, Elo difficulty, URL). It distinguishes from siblings like get_benchmark and search_models, but could explicitly contrast with exact-match retrieval.

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 description implies usage for searching benchmarks by name fragment, but does not provide when-to-use or when-not-to-use guidance, nor does it mention alternatives such as get_benchmark for exact matches.

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?

With no annotations provided, the description carries the full burden. It discloses return values (rank, Elo, URL) and the default fusion behavior across reasoning-effort settings, which adds meaningful context beyond the schema. However, it does not mention any potential limitations or side effects, but for a read-only search this is acceptable.

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 three short sentences that front-load the main purpose, then enumerate return values and default behavior. Every sentence adds information and there is no fluff or repetition.

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 search tool with three well-documented parameters and no output schema, the description covers the essential aspects: what it does, what it returns, and its default variant handling. It lacks guidance on when to use it relative to siblings, but the tool is simple enough that this is a minor gap.

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 all three parameters (query, limit, include_variants) are already documented. The description adds only minor context (e.g., 'fused across reasoning-effort settings') that is essentially a paraphrase of the include_variants schema description, so it provides little additional 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?

The description uses a specific verb ('Find') and resource ('ranked models'), and specifies the search dimensions (partial name or provider). It clearly distinguishes from sibling tools like get_leaderboard and search_benchmarks by focusing on model 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?

The description implies use when searching for models by name or provider but provides no explicit guidance on when to prefer this over sibling tools like get_leaderboard or get_model. No exclusions or alternative tool references are given, so usage context remains implicit.

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