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

Beyond the annotations (readOnly, idempotent, openWorld, non-destructive), the description defines what is actually measured (edge TTFB = DNS+TCP+TLS+first byte), the probe cadence (every 5 minutes, direct connections, no gateway), the response contents (region label, dataset timestamp, ranked rows), and the cost trade-off of omitting region (larger response). This is unusually rich behavioral context.

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 purpose and scope, and every sentence carries distinct information (use cases, sibling routing, metric definition, return shape, region omission). It is on the long side for a two-parameter tool, but no sentence is redundant.

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 an output schema present, return values need not be explained, yet the description still summarizes them helpfully. Combined with the metric definition, region omission behavior, and sibling routing, an agent has everything needed to call this 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?

Schema coverage is 100%, so baseline is 3; the description earns a notch above by adding the consequence of omitting region ('all four regions at once (larger response)'), which the schema omits, and by restating the human-readable region mapping. The 'limit' parameter is left to the schema, which is acceptable given full 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?

States a specific action (rank hosted AI inference API providers), the exact metric (measured network latency inside a probe region), the time window (last 24 hours), and the population size (46 tracked, with named examples). It explicitly distinguishes itself from sibling get_provider_latency, so an agent can route without opening 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?

Gives concrete triggering questions ('which provider responds fastest from Europe/US/Tokyo/Sao Paulo') and the decision rule to pick an endpoint near users. It names the alternative (get_provider_latency) and the condition that selects it (following one named provider across regions), and states exclusions (not tokens per second or model quality).

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 cover readOnly/idempotent/openWorld, but the description adds non-obvious operational context: hourly measurement from a single EU server and one account per vendor, the 24h-vs-7day window with Mann-Whitney significance flags, and the fixed hard-question accuracy panel. It also discloses the vantage-point limitation, which an agent needs to interpret results honestly.

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 dense but front-loaded sentences: purpose and metrics first, then the usage trigger, then the caveat and alternative. Every sentence carries weight, though the metric list is long enough to flirt with over-specification.

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 an output schema present and annotations covering safety, the description still supplies the measurement methodology, comparison window and statistical method needed to interpret the return values correctly. Nothing material to correct invocation 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 100% and the single enum parameter already documents both values plus the omit-for-both behavior. The description adds no parameter syntax beyond that, 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?

States a precise verb+resource ('Report whether the Claude Code or Codex CLI coding agents got slower, think less or answer worse than their own recent baseline') and enumerates the exact metrics returned. It also explicitly separates itself from sibling get_ai_api_latency for raw API latency.

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 phrase ('Use it when someone asks is Claude Code slow / nerfed today') and an explicit alternative condition ('for raw API latency use get_ai_api_latency'). This is exactly the when/when-not/alternative pattern.

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 cover the safety profile (readOnly, idempotent, non-destructive, openWorld), so the bar is lower. The description adds genuinely useful behavior beyond them: default scope is upcoming shutdowns only, include_past widens it, and entries carry a provider page they were verified against. It stops short of describing caching, freshness, or rate limits, so not a full 5.

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 sentences, front-loaded with what the tool returns, then usage, then the sibling disambiguation. Every clause carries information, though the long returned-field enumeration is dense enough that it approaches a data dump.

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 two-parameter read-only lookup with a full output schema and complete annotation coverage, nothing an agent needs is missing: scope defaults, filter behavior, replacement guidance, and sibling routing are all present.

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 both parameters are already documented with valid provider values and the include_past default. The description restates the include_past effect ('also get already-retired models') and omits provider filter syntax, adding little beyond the schema. 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 (list) and resource (announced AI model shutdowns), enumerates the fields returned (announcement date, shutdown date, replacement, verification page, notice statistics), and explicitly distinguishes itself from the sibling latency tools. An agent can identify this tool without opening the 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 explicit when-to-use triggers ('before pinning a model version, to plan migrations, or to check whether a model ID still works') and names the alternative for a different need ('for speed use get_ai_api_latency'). Both the selection condition and the exclusion are 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, openWorldHint and destructiveHint=false, so safety is covered structurally. The description adds meaningful behavior beyond them: the 24-hour data window, the enumerated probe regions, the exact metric set, and the error contract for an unknown provider id. Return format is not detailed, but an output schema exists to carry that.

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 sentences, front-loaded with what the tool returns, then examples, then the alternative tool. No filler, no restatement of the title, and each sentence carries distinct routing or scoping 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?

An output schema exists so return values needn't be explained, yet the description still names the metrics and regions the agent will see. Combined with usage examples, the sibling routing rule, and the error behavior, everything needed to select and call this single-parameter tool correctly is present.

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 format is already pinned by the schema (lowercase-with-hyphens, example ids), so the baseline is 3. The description adds a semantic the schema does not: an unknown provider id errors and lists the valid ids, which tells the agent how to recover from a bad value.

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?

Specific verb+resource: 'Show how a single AI API provider performs from every probe region', with the exact metrics returned (rank, median and p95 TTFB, uptime, sample count) and the time window (24 hours). It explicitly contrasts scope with the sibling get_ai_api_latency, which is the full leaderboard, so an agent can distinguish them without opening 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?

Provides concrete example questions ('is the Anthropic API slower from Asia than from Europe', 'was OpenAI reachable today') that map intent to this tool, and names the alternative get_ai_api_latency plus the condition that selects it (full region leaderboard). When-to-use and when-not-to-use are both covered.

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