Local AI Prices

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

Annotations already declare a safe, idempotent, closed-world read, so the description doesn't need to reargue safety. It adds real behavioral context beyond them: results are limited to tracked machines, filtered by a 4-bit memory-fit rule, sorted cheapest-first, and include measured tok/s only where a result has been published.

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 tight sentence with the important scoping (tracked machines, fit rule) front-loaded and ordering stated before the speed caveat. The nested parentheticals make it dense, though no sentence 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 read-only lookup with no output schema and fully documented parameters, the description covers the essentials: filtering, ranking, and what the speed figure represents. It could say more about what each returned machine record contains, but nothing critical to invoking it correctly 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 model, region, max_price, measured_only and limit are already documented in the schema. The description adds only a small amount of meaning beyond that (the model id can be a name, and ids come from list_models); it does not clarify the currency behavior of max_price or the implicit-cap behavior of limit beyond what the schema 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?

Names a specific resource (tracked machines that can run a given open model), a concrete fit criterion (4-bit weights in memory), and an ordering rule (cheapest first). It also distinguishes itself from list_models by deferring id lookup to that sibling, so an agent can separate the two without opening schemas.

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?

Explicitly routes the agent to list_models when it lacks a model id, which is the main prerequisite. It does not contrast itself with search_machines or get_machine, so the when-not-to-use case is left implicit rather than stated.

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, idempotentHint, non-destructive and closed-world, so the safety profile is covered. The description adds real behavioral value by enumerating what comes back: price per store with stock and check time, price history summary, every published speed result with its source, and value. It stops short of saying what happens on an unknown id, but the payload contract is far clearer than the annotations alone.

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 tightly packed sentence with the identity statement front-loaded and the returned-fields list following. No filler, no repetition of the title, and nothing that 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 single-parameter read tool with no output schema, the description carries the missing return-value contract by listing exactly which data blocks the agent will get. Combined with annotations covering safety and idempotency, nothing essential for correct invocation is absent.

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 the id parameter already documented including the URL form and an example, so the baseline is 3. The description adds marginal but genuine meaning by tying the id to search_machines as its origin and restating that a page URL is equally acceptable.

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?

Opens with a specific verb and resource: fetch 'one machine in full by id' or page URL. It also names the sibling search_machines as the source of the id, so an agent can tell it apart from the search tool without inspecting either 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?

The parenthetical '(from search_machines)' establishes the intended workflow: search first, then fetch details with the resulting id. There is no explicit when-not-to-use or mention of an alternative for bulk lookups, but the retrieval path is 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?

Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false and a closed-world scope, so the safety profile is covered. The description still adds useful behavior: newest-first ordering, cursor-style pagination via next_since, and a daily refresh cadence. It does not describe the item shape or how 'limit' interacts with the cursor, leaving gaps for a tool with 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?

Three short, front-loaded sentences: what you get and in what order, then how to page, then a freshness caveat. Nothing is redundant, though the freshness note and pagination note could be compressed slightly.

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 and zero required parameters, the description carries the return-shape burden and does cover ordering and the next_since field. It stops short of describing the actual change record contents or how limit constrains the page, which an agent would still have to discover.

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 description goes further by explaining how to source the 'since' value (a timestamp, or the previous reply's next_since), which is the non-obvious part of the parameter that the schema doesn't convey.

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 the resource (changes) and enumerates the exact kinds returned (price drops, price rises, restocks, newly tracked machines), plus result ordering 'newest first'. That enumeration matches the types enum in the schema and makes it obvious this is a change feed, not a machine/model lookup like its siblings.

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 concrete usage conditions: set 'since' to a timestamp to bound the window, reuse 'next_since' from the reply for the next call, and it warns that prices are only checked once a day so results can lag. No explicit when-not guidance or sibling routing is given, but no sibling overlaps this feed.

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, idempotentHint, openWorldHint=false and destructiveHint=false, so the safety profile is fully covered by structured data. What the description adds is the return shape — far more useful than repeating 'read-only' — but it does not say whether the catalog is paginated, whether the list is exhaustive, or how 'how many machines fit each' is derived. With annotations doing the behavioral work, a 3 is right.

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 sentence, front-loaded with the resource, then a compact enumeration of returned fields. There is no filler, no restated title, and no marketing language. Every clause conveys a distinct field the agent will receive.

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 describing return values, and it largely does so by listing the fields. That is sufficient for a zero-parameter read tool. Minor residual gaps: it does not clarify pagination, ordering, or what 'how many machines fit each' means quantitatively.

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 schema coverage is trivially 100% and there is no parameter meaning to add. Baseline 4 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 names the resource unambiguously — 'the open models we track' — and enumerates the attributes returned (id, size, memory at 4-bit/8-bit, machine fit counts, measured-speed count, page URL). Combined with the title 'List models', an agent knows exactly what this tool yields. It does not, however, distinguish itself from the sibling find_for_model, which is model-related, so a small differentiation gap remains.

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: a zero-parameter enumeration of the model catalog naturally reads as 'call this to get the full list.' There is no explicit statement of when to prefer it over find_for_model or the other siblings, nor any exclusion such as 'use find_for_model when you want machines that can run a given model.' That is adequate but leaves routing to inference.

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 a safe, read-only, idempotent, closed-world operation, so the bar is lower. The description still earns credit by disclosing a real behavioral trait beyond them: prices are refreshed only once per day, so results may be stale, and it lists the exact metrics returned (price, tok/s, value, check date, URL).

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, front-loaded with purpose and immediately followed by return contents and the freshness caveat. Nothing is repeated and 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 an eight-parameter, all-optional search with no output schema, the definition covers purpose, returned fields, and data-freshness limits adequately. It stops short of explaining result ordering behavior or what an empty result means, which would have made it fully self-contained.

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 the enums already document defaults and meanings for kind, sort, region and benchmark. The description only restates the same axes (budget, memory, type) without adding units, edge cases, or interaction notes between parameters such as max_price and sort=price, so 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?

States a specific verb and resource ('Find local AI computers and graphics cards') plus the three filtering axes (budget, memory, type), and enumerates what is returned. It does not differentiate itself from the closest sibling, find_for_model, so an agent must infer the split between them.

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 filtering dimensions ('by budget, memory and type'), which tells the agent this is a discovery/browse tool rather than a lookup. There is no explicit when-to-use statement, no mention of when to prefer find_for_model or get_machine, and no stated prerequisites.

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