GPU Cloud Prices (AWS, Azure, OCI)

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

With no annotations, the description carries the full burden and does so exceptionally: it discloses data freshness (cached snapshot with as_of, not live), the payment requirement ($0.005 USDC on Base/eip155:8453), the exact failure mode (unpaid calls return HTTP 402 with PAYMENT-REQUIRED) and recovery (retry with PAYMENT-SIGNATURE), plus that tools/list is free. This is precisely the auth, cost, and staleness context an agent needs.

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 opening sentence is keyword-stuffing filler ('H100 price per hour, A100 price per hour, A100 rental price, Azure Spot GPU price') that the following sentence already covers more cleanly, so it does not earn its place and front-loads noise rather than purpose. The payment and freshness details are necessary and tightly written, but the redundancy costs 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?

With no output schema and no annotations, the description does a strong job covering cost, freshness, and provider coverage, and it hints at the return shape ('price lists with as_of'). It stops short of describing pagination/limit behavior or the default sort ordering, which the absent output schema would otherwise have to supply.

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 only 40% across 10 parameters, so the description should compensate more. It implicitly touches the sort dimension (hourly vs per-GPU-hour), the provider option, the offer_type option, and the query dimension (GPU models), but leaves region, min_gpus, include_fractional, limit, and the max_price filters unexplained. Partial but incomplete compensation.

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 a specific resource (hourly and per-GPU-hour prices for data-centre GPUs) and scopes it precisely across GPU models (H100, H200, A100, L4...), providers (AWS, Azure, OCI), and offer types (on-demand, Azure-only Spot). An agent immediately knows this is a GPU price lookup, and the sole sibling (uk_vat_calc) is unrelated so no disambiguation is needed.

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 clearly frames when the data is appropriate by stating it is a 'Cached snapshot... not live' and noting the Spot availability limitation (Azure-only in the current snapshot), which tells the agent when results may be incomplete. It also spells out the invocation path (payment flow for tools/call). There are no competing alternatives to exclude, but no explicit exclusion language either.

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 disclosure burden and does so well: it explains fallback behavior (OCR, headless Chrome), the paid-call flow ($0.001 USDC on Base, HTTP 402 + PAYMENT-SIGNATURE retry), that failed fetches are not charged, and the OCR page cap. These are exactly the non-obvious traits an agent needs before invoking.

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?

Dense but front-loaded and well-parenthesized; the core purpose leads and payment constraints follow. It is long, but nearly every clause (pricing, 402 retry, free failures, deep alternative) carries operational weight, so little 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?

An output schema exists, so return-shape explanation is a bonus rather than a requirement, and the description still names the returned fields. Combined with the auth/payment model, fallback behavior and endpoint alternatives, an agent has everything needed to call it correctly.

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?

At 57% schema coverage the description compensates meaningfully: it clarifies what the format options yield (clean markdown vs text/html), explains the OCR auto behavior, and adds the OCR first-3-pages limit that isn't in the schema. max_chars, include_links and ocr_lang still get no descriptive coverage anywhere, so it falls short of full compensation.

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+resource ("fetch a public URL and return the main content as clean markdown") and names the outputs (title, lang, dir, word_count, sha256). An agent immediately knows this is a page-fetching/extraction tool, distinct from the pricing/VAT/verification 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?

Gives clear operating conditions: OCR auto-triggers for scanned PDFs or garbled text layers, headless Chrome is used for JS-rendered pages, robots.txt is respected, and it names an alternative endpoint (/v1/read/deep) for 10-page OCR. It does not state when NOT to use the tool or contrast it with the siblings, but the context is otherwise unambiguous.

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 important context: the x402 payment requirement, exact cost, HTTP 402 flow, and PAYMENT-SIGNATURE retry — plus a 'Not tax advice' disclaimer. However, it says nothing about how conflicting inputs (net vs gross vs lines) are resolved, defaults for rate/kind, or failure behavior, which are significant for a 9-param calculator.

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?

Purpose is front-loaded, and the dense payment sentence is necessary operational content rather than filler. The 'Not tax advice' tail is minor but harmless. Overall it is tight with no wasted sentences.

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 tool with 9 parameters at 0% schema coverage, no annotations, and no output schema, the description covers only purpose and payment mechanics. It omits parameter semantics, default values, input precedence, and any notion of what the calculation returns, leaving the agent under-informed before invocation.

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 0% across 9 parameters, so the description must compensate. It hints at net, gross, lines, and threshold, but leaves kind, rate, markdown, description, and monthly_taxable_turnover entirely unexplained, and gives no format or precedence rules (e.g., whether net and gross are mutually exclusive).

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 ('UK VAT calculator') and enumerates its scope: net/gross, multi-line, £90k threshold. This clearly identifies the tool's function. It doesn't need to differentiate from siblings since the only sibling (get_gpu_prices) is unrelated, but the differentiation is implicit rather than stated.

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 guidance on when to use this tool, when not to, or how to choose between the net/gross/lines input modes. The operational note ('Same as POST /v1/vat-calc') and payment mechanics are about invocation, not usage selection. An agent gets no help deciding whether this tool applies to its task.

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 burden and does so well: SSRF guarding, robots.txt honoring, headless Chrome fallback, paid x402 flow with 402/PAYMENT-SIGNATURE retry, and the exact conditions under which no charge occurs after an upstream failure.

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?

Front-loads the purpose before mechanics and payment, and every clause is informative. It is dense and long, with the enumerated return fields arguably duplicating the output schema, but nothing is padding.

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 paid, no-annotation tool with six parameters and an output schema, the description covers purpose, security posture, extraction behavior, billing semantics and error non-charge cases. An agent has everything needed to call it and handle the 402 response.

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 83%, so the schema already documents url, path, render and tolerance. The description adds only modest extra semantics such as numeric tolerance matching and string trimming, plus an implicit mapping of mode to extraction type, which is near the baseline for high-coverage schemas.

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 precise verb+resource: server-side fetch of a live URL and comparison of an extracted value against a claim, with the extraction mechanisms named. An agent can distinguish this from read_url (raw fetch) or x402_conformance without inspecting 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?

Usage is implied by the name and description (verify a claim about a live page), and the payment retry flow is spelled out, but there is no explicit when-to-use-this-vs-sibling guidance, e.g. why verify_value rather than read_url plus local comparison.

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 so: it states the tool never signs or sends a payment, names the exact price and chain ($0.03 USDC on Base, eip155:8453), documents the 402 challenge/retry handshake, and specifies 400/424 non-charge outcomes. This is unusually complete for a mutation/payment-gated 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?

Front-loaded with the core action, and the long check list is dense with distinct evidence rather than filler. It is verbose, but nearly every clause (check names, status codes, pricing) earns its space.

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?

An output schema exists, so return-value structure needn't be explained, yet the description still conveys the 0-100 grading and per-check evidence format, plus the full payment and pricing flow. Nothing an agent needs to invoke 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 description coverage is 67%, so most parameters are already documented in the schema. The description only obliquely touches parameters, noting POST-body behavior via 'sent with method POST on every unpaid probe' and the public http(s) 80/443 constraint, without adding format or semantic detail 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?

States a specific verb and resource ('probes an x402-paywalled URL UNPAID ... grades it 0-100 with per-check pass/warn/fail evidence') and enumerates the exact checks performed. It is unmistakable against siblings like read_url or verify_value, which share no domain.

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?

Explains the paid workflow clearly: unpaid calls return HTTP 402 with PAYMENT-REQUIRED and the agent must retry with PAYMENT-SIGNATURE, and it states the three conditions under which no charge occurs. What is missing is any explicit routing guidance against alternative tools, though no sibling overlaps this function.

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