SpreadBench NCSX — AI Infrastructure Bond Credit Spread Index

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

With no annotations, the description must carry the behavioral burden, and it partly does: the verb 'Get' signals a read operation and the parenthetical discloses units (basis points) and the nature of the content. It does not state freshness/caching, whether the attribution covers only top contributors or the full index, or any auth or rate-limit constraints.

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 with the resource and time window front-loaded, and the parenthetical earns its place by decoding what 'attribution' returns and in what unit. Nothing is redundant or padded.

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-argument, no-output-schema tool the description covers the essentials: what it fetches, the time period, the unit, and the conceptual return shape. It stops short of describing the return structure (ranked list, sign convention, number of contributors), which is the one thing an agent cannot recover from the empty schema.

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 there is nothing for the description to disambiguate beyond what the empty schema already shows. Baseline 4 applies; no parameter meaning is missing.

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 pairs a specific verb ('Get') with a specific resource ('this week's index change attribution') and clarifies the payload with a parenthetical ('which bonds moved the index, in bp'). That distinguishes it in substance from siblings like ncsx_levels or ncsx_constituents, though it never names those siblings to route the agent 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: 'this week's' fixes the time window to the current week, so an agent knows the scope without supplying arguments. There is no explicit statement of when to prefer this over ncsx_weekly, ncsx_levels, or ncsx_bond, and no exclusions or prerequisites are 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?

With no annotations, the description carries the full behavioral burden, and it does disclose the return content (profile plus a 30-day spread path), which is genuinely useful. However, it says nothing about failure behavior when the CUSIP is not an index constituent, data freshness, units/frequency of the spread path, or any access constraints.

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 with the verb, scope, and key parameter front-loaded and no wasted wording. Nothing is buried or repeated.

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 tool this is nearly adequate, and the description at least characterizes the returned data. With no annotations and no output schema, though, an agent still lacks the return structure, spread-path units/tenor convention, and behavior for non-constituent CUSIPs.

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 schema already defines the 9-character CUSIP pattern with the identical example, so the description adds no meaning beyond it. Baseline 3 is appropriate for a single fully-documented parameter.

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 gives a specific verb ('Get') and a precisely scoped resource: the profile and 30-day spread path of a single constituent bond, keyed by CUSIP. It is clearly distinguishable from index-level siblings like ncsx_levels or ncsx_weekly, though it never names those siblings 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 use case is implied by 'one constituent bond by CUSIP' — drill into a single name rather than an index aggregate — but there is no explicit when-to-use versus alternatives and no stated preconditions. An agent can infer the routing, but it must do the inferring itself.

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 carries the full behavioral burden. 'List all' implies a bulk retrieval with no filtering, but it says nothing about pagination, dataset size, caching, or freshness, and it mentions 'identifier verification status' without explaining what that status means or where it comes 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?

A single front-loaded sentence with no padding. The parenthetical field list is compact and informative, though 'identifier verification status' is appended without explanation and slightly clutters the sentence.

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-parameter list tool there is little to get wrong, and the field inventory helps. But with no annotations and no output schema, the description should say more about result shape and scale (is this the whole index? how large? how fresh?) to let an agent call it confidently.

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 there is nothing to disambiguate and the baseline is 4. The description instead documents the fields that come back, which is useful orientation even though it is not a parameter concern.

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 gives a specific verb ('List all') and a specific resource ('constituent bonds'), and it enumerates the fields returned (issuer, coupon, maturity, rating, tier, spread) plus a verification-status attribute. It doesn't distinguish itself from siblings like ncsx_bond or ncsx_latest, but the 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?

There is no when-to-use guidance at all: no mention of when a full constituent listing is preferable to the sibling tools ncsx_bond, ncsx_latest, or ncsx_levels. An agent must infer the choice from the name alone.

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 and no output schema, the description carries the full burden and does well to enumerate the exact return payload (levels, daily/weekly changes, 90-day percentiles, as-of date, methodology version, DOI, data_status). The data_status caveat discloses a real behavioral trait about data provenance that is not derivable from any structured field. Missing: it never states the tool is a read-only, side-effect-free call, nor what happens when data is unavailable or stale.

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 filler, and the return contents are front-loaded before the quoting caveat. The payload enumeration is dense but every item listed is information the agent cannot get elsewhere since no output schema exists.

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-parameter read tool with no output schema, listing the returned fields and the data_status interpretation rule covers the essentials an agent needs before quoting numbers. The remaining gap is routing guidance against the five sibling tools and any note on failure or staleness behavior.

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 there is no parameter semantics to explain and the baseline of 4 applies. The description correctly does not waste space on nonexistent arguments.

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?

"Get the latest NCSX index levels (Composite / HY / IG)" gives a concrete verb plus resource and even enumerates the sub-series returned, which is far beyond a title restatement. It distinguishes itself from ncsx_levels and ncsx_weekly by the word "latest" plus the bundled daily/weekly changes, but it never explicitly names those siblings as the alternative for historical data.

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 one explicit instruction — "Read data_status before quoting" with the seed/verified semantics — is genuinely useful conditional guidance. However, there is no statement of when to pick this over ncsx_levels or ncsx_weekly, leaving the agent to infer that "latest" means the current snapshot rather than a time series.

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 carries the full burden. It usefully discloses the window length (90 trading days) and granularity (daily), which the schema does not. It says nothing about return shape, ordering, error behavior for an invalid tier, or whether the series is point-in-time level only, leaving real gaps for a data-retrieval 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?

A single front-loaded sentence with zero filler; the scope constraint and the parameter domain both appear before any trailing detail. 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?

This is a simple one-parameter read tool with no output schema and no annotations, and the description adequately conveys the returned content (daily level series over 90 trading days for a single tier). It stops short of describing ordering or units, which is a minor residual gap at this complexity level.

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 tier parameter already has an enum with a description, so the schema does the heavy lifting. The description merely restates the enum values in parentheses without adding meaning such as what distinguishes hy from ig beyond the name.

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 pairs a specific verb ("Get") with a precisely scoped resource ("90-trading-day daily level series") and names the tier parameter's allowed values. The 90-day daily window implicitly separates it from ncsx_latest and ncsx_weekly, but 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 implied by the scope: an agent can infer this is the tool for a historical daily series rather than the latest snapshot. However, there is no explicit statement of when to prefer ncsx_latest, ncsx_weekly, or ncsx_bond, and no exclusions or prerequisites are 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?

With no annotations, the description must carry the behavioral burden; "Get" signals a read-only, side-effect-free operation, and the field list hints at the return contents. However, it doesn't state that the response is static/versioned, whether it's safe to call repeatedly, or anything about result format, which is the main gap for an unannotated 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?

A single front-loaded sentence with the verb and resource first, followed by a compact field list. Every clause adds information and nothing is repeated or padded.

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-parameter metadata endpoint with no output schema, the description usefully enumerates what comes back, which substitutes for the missing output schema. It stops just short of explaining how those values should be used (e.g., for citation or compliance checks).

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 there is nothing for the description to disambiguate and the baseline of 4 applies. The description correctly implies a parameterless call by describing a single fixed metadata payload.

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?

"Get methodology metadata" names a specific verb and resource, then enumerates the exact fields returned (version, DOI, license, formulas, licensing status). This clearly separates it from the sibling tools, which all appear to return numeric data (bond, levels, weekly, latest) rather than documentation, though it doesn't explicitly state that contrast.

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 you call this to cite sources, check licensing, or look up formulas. There is no explicit statement of when to prefer it over ncsx_attribution or ncsx_latest, and no mention of caching or frequency expectations for what is clearly static metadata.

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 supplied, so the description carries the full behavioral burden. It does disclose the shape of the returned snapshot, which is the most valuable disclosure for a read tool like this, but it says nothing about whether the snapshot is cached, how 'latest' is resolved relative to the current date, or what period the weekly change spans.

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 with the resource named first and the returned fields compactly parenthesized. Nothing is redundant and nothing is buried.

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-parameter tool with no output schema, the parenthetical field list does useful work standing in for return documentation. However, with six closely named siblings, the definition leaves an agent unable to decide between this and ncsx_latest or ncsx_levels, which is the main completeness 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?

The tool takes zero parameters, so there is no per-parameter meaning to add and the baseline of 4 applies. The word 'weekly' does imply a fixed reporting period rather than a caller-supplied window, which is consistent with the empty 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?

Uses a specific verb ('Get') plus a specific resource ('the latest weekly report snapshot') and enumerates the payload (levels, weekly changes, attribution, citation string). The purpose is unambiguous, but the description never distinguishes the tool from heavy sibling overlap — notably ncsx_latest and ncsx_levels, whose names suggest they may return much of the same content.

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 or when-not-to-use guidance, and none of the six siblings are named as alternatives. At best an agent can infer 'use this when you want a bundled weekly view,' which is thin given that ncsx_latest and ncsx_levels plausibly compete for the same request.

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