if-the-card-fits

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

No annotations are provided, so the description carries the behavioral burden. It discloses that the comparison is grounded in the person's own spend and reported per year, but it says nothing about the response shape (no output schema exists), whether the call is read-only, or that the 'source' slug is used for click attribution — context that lives only inside the schema. Adequate, not complete.

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 with zero filler: the operation and its result come first, the routing rule second. Every clause earns its place 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 an 11-parameter tool with nested objects, no annotations, and no output schema, the description covers intent but leaves the return shape and the read-only nature implicit. It also doesn't mention the prerequisite of mapping spending via spending_categories, which only appears in the totals schema text.

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%, including the non-obvious constraints on totals and the years enum, so the schema does the heavy lifting. The description adds a framing ('two named cards', 'on the person's own spending', annual horizon) but no per-parameter meaning beyond what the schema already says.

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 first sentence names the exact operation — two named cards compared head to head on the person's actual spending — and states the output (which pays more, annual dollar delta). It also explicitly separates itself from simulate_cards, so the agent can distinguish it from siblings without opening a schema.

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 an explicit trigger ('when they are choosing between two specific cards') and an explicit alternative with its own trigger ('what should I get' → simulate_cards). This is the when/when-not/alternative pattern at full strength.

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 burden. It usefully clarifies that the tool is advisory ('says whether… and what to move to') and states the return shape ('a verdict and a rounded yearly margin'), but it omits safety profile, side effects, the required attribution source, and limits such as max held cards or data 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?

Two sentences, zero waste, and the purpose is front-loaded. The return-value note is compact and integrated naturally.

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 9-parameter tool with nested objects and no annotations, the description covers the core purpose and return shape but leaves important gaps: it does not disambiguate from sibling tools, does not mention required 'source' attribution, and does not state the advisory-only nature as explicitly as a no-annotation tool should.

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 already explains all nine parameters thoroughly. The description adds only the redundant instruction 'Pass the cards as held' and a note about return values, without adding meaning beyond what the schema provides.

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 states a specific function: judging whether held cards still justify their annual fee and suggesting alternatives. It implies differentiation from siblings by focusing on 'cards the person already carries,' but does not explicitly name compare_two or simulate_cards to rule them out.

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?

It gives implied context ('cards the person already carries') and a direct instruction ('Pass the cards as held'), but offers no explicit when-to-use, when-not-to-use, or named alternatives among the sibling tools. The agent must infer that this is for existing-card evaluation rather than comparison or simulation.

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 burden and does well: it discloses provenance (model checked to the cent against an independent calculator), that card terms were human-verified against issuer pages, and that the response reports catalog version and last-checked dates. It also discloses a side effect (apply clicks attributed to the source slug) via the required source parameter. It does not state rate limits or confirm the operation is read-only/no destructive side effects.

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?

Four sentences, front-loaded with what is returned, followed by provenance and the two display obligations. Every sentence carries information; only the sourcing sentence is somewhat dense but it justifies itself by establishing trust in the numbers.

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 9 parameters, nested objects and no output schema, the description does the heavy lifting by enumerating the return contents: top three cards, best two-card combo, post-fee value over the horizon, catalog version, last-checked dates, continueUrl and disclosure. That is enough for an agent to know what comes back, though exact field shapes remain unspecified.

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 already documents every parameter, including the totals category constraint and the years enum. The description adds little parameter-level detail beyond 'a year of spending by category' and 'the chosen horizon', 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 gives a specific verb and resource: given a year of spending by category, it returns the three best cards plus the best two-card combination valued after annual fee. That is concrete and unambiguous. It does not, however, distinguish itself from siblings compare_two or keep_or_cancel, which an agent would need to route correctly.

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?

It names a prerequisite workflow by pointing to the spending_categories tool to map the statement onto categories first, and it prescribes two mandatory follow-up actions (show continueUrl, show disclosure string). What is missing is any explicit guidance on when to prefer compare_two or keep_or_cancel instead of this simulation.

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 load; it discloses the tool's role as the id-authority that simulate_cards consumes, which is the key behavioral fact. It omits whether the 19 categories are a fixed constant and what fields each entry carries, but the workflow relationship is well communicated.

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, nothing wasted: the first defines the resource, the second gives the sequencing instruction. The ordering guidance is front-loaded where an agent will act on it.

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 catalog with no output schema, knowing there are 19 entries and that their ids feed simulate_cards is essentially sufficient to call it correctly. A brief note on the shape of each entry (id plus label) would close the only remaining 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 nothing for the description to disambiguate; the baseline for a no-param tool applies. The description spends its space on how to use the returned ids rather than mis-describing inputs.

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?

Names a specific resource — the 19 spending categories the API scores against — so the agent knows it's a static catalog lookup, not a computation. It's clearly distinct from compare_two, keep_or_cancel and simulate_cards, though it never states outright that it takes no arguments and returns a fixed 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?

"Call this first" gives explicit ordering, and "map the person's statement onto these ids before calling simulate_cards" names the exact downstream dependency and the prerequisite step. No inference is required about when to invoke it or what to do with the result.

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