14 models are statistically tied for #1 in Arena math, spanning 49 points. 4 are Flash models.
Scored across 3 tools
Each tool serves a clearly distinct purpose: searching charts, retrieving a specific chart's full data, and fetching the daily chart feed. There is no overlap in functionality, and the descriptions make the boundaries obvious. An agent can easily select the right tool for a given task.
All tool names follow a consistent verb_noun pattern in snake_case: get_chart_data, get_venture_chart_of_the_day, and search_charts. The verbs are appropriate for the actions (get vs. search), and the naming is predictable. No naming inconsistencies are present.
Three tools are well-scoped for a server that provides access to a finite set of research charts. Search enables discovery, get_chart_data retrieves details, and get_venture_chart_of_the_day provides a daily feed. No tool feels redundant or missing in count.
The surface covers search, retrieval, and daily updates, which are the core operations for accessing charts. However, there is no explicit tool to list all charts, though search_charts can serve that purpose with an empty query. This minor gap is easily worked around.
Retrieves the complete empirical data series, takeaway, notes, and primary sources for a specific Mode VC chart by its slug or title.
| Name | Required | Description | Default |
|---|---|---|---|
| slug | Yes | Chart slug or identifier (e.g. "pre-seed-valuation-dispersion", "ai-inference-cost-comparison", "open-source-vs-closed-ai-models-gap") |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnly, idempotent, non-destructive, and closed-world, so safety is covered. The description adds value beyond that by enumerating what the response contains (data series, takeaway, notes, primary sources) and confirming it is a per-chart lookup, though it omits behavior on an unresolved slug/title.
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 front-loaded sentence with no filler; the identifier and the returned payload are stated up front without redundancy.
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 correctly compensates by enumerating the returned fields, which is what an agent needs for this simple one-parameter lookup. Only error/ambiguity behavior for non-matching or title-based identifiers is left unstated.
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 example slugs, so the schema already carries parameter meaning. The description's 'slug or title' phrasing adds a hint about accepted identifiers but subtly conflicts with the schema, which defines only a required 'slug' string.
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 (Retrieves) plus the exact resource contents (empirical data series, takeaway, notes, primary sources) scoped to a Mode VC chart identified by slug/title. It is clearly distinguishable from get_venture_chart_of_the_day and search_charts by content, though it does not name those alternatives.
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 tool fetches full detail for a chart you must already be able to identify, which suggests it follows a search/discovery step (search_charts). No explicit when-to-use, when-not-to-use, or alternative is stated.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Returns the latest daily AI adoption, developer trends, model benchmark movements, and pricing deflation charts from Mode VC's primary research lab.
| Name | Required | Description | Default |
|---|---|---|---|
| limit | No | Number of recent charts to retrieve |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true, idempotentHint=true, destructiveHint=false, and openWorldHint=false, so the safety and idempotency profile is fully covered structurally. The description adds useful content about what subjects the charts cover (AI adoption, developer trends, model benchmarks, pricing deflation), but says nothing about return format, ordering, or how many charts are actually available for the requested days.
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 front-loaded sentence with no filler; the key noun phrase ('latest daily ... charts') arrives early. It is tightly sized, though the list of chart topics is somewhat enumerative rather than maximally economical.
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 partially compensates by enumerating the chart topic categories, which is genuinely helpful for anticipating what comes back. However, for a tool whose purpose is to retrieve a specific daily artifact, it omits how results are structured, whether one chart per topic per day is returned, and how the limit interacts with the topic categories.
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 there is only one optional parameter with a default (limit=3, max=10), so the schema fully documents it. The description adds nothing about the limit parameter, making it baseline 3 by the rule for high coverage.
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 has a clear verb ('Returns') and a specific resource ('latest daily ... charts') scoped to Mode VC's own research lab output. The 'daily' framing implicitly distinguishes it from the sibling search_charts and get_chart_data, but it never names those siblings, so the differentiation is left to inference.
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 explicit when-to-use guidance, no prerequisites, and no mention of the alternatives (search_charts, get_chart_data) that also return chart content. The only cue is 'latest daily', which implies a freshness-based use case but does not tell the agent when this tool should be preferred.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Searches across all 38 empirical AI & venture research charts published by Mode VC. Filter by keyword, topic, or date. Returns titles, empirical claims, takeaways, canonical URLs, and markdown endpoints.
| Name | Required | Description | Default |
|---|---|---|---|
| limit | No | Maximum number of charts to return | |
| query | No | Search term matching title, claim, key takeaway, or keywords (e.g. "pre-seed", "pricing", "claude", "inference", "sdk") |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false, and openWorldHint=false, so safety is covered. The description adds useful behavior by disclosing the returned fields (titles, empirical claims, takeaways, canonical URLs, markdown endpoints), which matters because no output schema exists. Minor gap: no pagination/result-limit behavior beyond the schema's limit.
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 tightly packed sentences with zero filler, leading with the scope and action before listing filters and return fields. Every clause carries 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?
With no output schema, the description correctly picks up the burden by enumerating what is returned, and the annotation set covers the safety profile. It is nearly complete but leaves the relationship to sibling retrieval tools (e.g., get_chart_data) unstated.
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 fully documents query and limit, establishing a baseline of 3. The description adds little parameter meaning and actually introduces a discrepancy: it advertises filtering "by date" although no date parameter exists in the schema, which could mislead an agent.
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") and resource ("38 empirical AI & venture research charts published by Mode VC"), with a concrete count that bounds scope. It does not explicitly differentiate itself from siblings like get_chart_data or get_venture_chart_of_the_day, so an agent must infer the routing, but the core purpose is unambiguous.
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 implied usage ("Filter by keyword, topic, or date") but gives no when-to-use-vs-alternative guidance and no exclusions against the sibling retrieval tools. An agent can guess that this is the discovery step before get_chart_data, but nothing in the text says so.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
get_chart_dataget_venture_chart_of_the_daysearch_chartsThe gap between Arena's #1 and #10 models fell from 148 points in Feb 2024 to 18. #10 now wins.
Arena rates 399 text models from 52 labs, up from 10 models in May 2023. Yet only 5 labs hold.
The best open model trails the best closed model by 42 Arena points. In a blind vote it would.
Since Jan 2025, Arena's #1 spot has changed hands 16 times. Half of the finished reigns lasted.
Add one secure layer between your agents and this server.