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

The description repeatedly states 'Read-only' and 'ADVICE ONLY. nable never applies the action,' but the annotations declare readOnlyHint: false (i.e., not read-only). This is a direct contradiction between the description and structured metadata. The tool's safety profile is misrepresented, so the agent cannot reliably assess 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?

The description is longer than necessary but well-structured with line breaks, bullet-like verdicts, examples, and trigger phrases. It front-loads the core purpose and each section adds useful context. Minor redundancy exists ('ADVICE ONLY' appears twice), but overall it remains organized and readable.

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?

The tool has 7 parameters and no output schema, so the description needs to explain behavior. It covers the verdicts, policy knobs, and action_type examples, making the tool's behavior understandable. It does not detail the exact meaning of every parameter (handled in parameter_semantics) and omits response format, but for an advisory policy gate, the description is largely adequate.

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?

With schema description coverage at 0%, the description must compensate. It explains the role of 'action_type' and mentions optional cost inputs (Terraform plan, helm diff, monthly delta), but it does not map these to the parameter names like tf_dir, terraform_plan_file, terraform_plan_json, or budget_name. This adds partial value but leaves several parameters ambiguous.

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 clearly identifies the tool as an 'Advisory policy gate' that returns a verdict (allow/escalate/block) for proposed remediation actions. It uses specific verbs and resource names, and the examples of action types and triggers distinguish it from siblings like estimate_change_cost or check_ai_budget.

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 description provides explicit good triggers ('can the agent do X', 'is this action within policy') and concrete example queries. It clearly indicates this is the propose-only check before a human applies an action. It does not explicitly name alternatives or say when not to use this tool, but the context is strong enough to infer appropriate usage.

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=true and destructiveHint=false. The description adds that it is 'Advice only, it never blocks' and describes the return shape (verdict, reason, recommendation), which goes beyond the annotations. This provides useful behavioral context for the agent.

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 that are all essential and well-structured. It starts with the primary purpose, then gives usage timing, output shape, and parameter semantics without any fluff.

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 simple tool with one optional parameter, no output schema, and clear annotations, the description fully covers purpose, usage, behavioral nature, output format, and parameter meaning. Nothing important 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 0%, so the description must compensate. It explains the parameter's purpose: 'Pass estimated_next_tokens to test whether the next task would tip a token budget over.' This gives clear meaning to the single optional integer.

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 clearly states the tool is an 'Advisory gate' for checking whether the agent is about to exceed its AI budget. It specifies the resource (AI budget) and the action (check), and distinguishes itself from sibling tools like get_cost_summary or estimate_change_cost by focusing on a pre-run verdict.

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 says 'Call this before an expensive run,' providing clear context for when to use it. It also instructs to 'relay the verdict and let the human decide,' but does not mention when not to use it or name alternative tools, so it falls short of a 5.

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 indicate readOnlyHint=false and destructiveHint=false, but the description goes beyond by explaining the propose-then-confirm flow, that it never creates/modifies/deletes anything in the AWS account, and that credentials stay local. This adds significant behavioral context beyond the annotations, covering safety and data handling.

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 description is concise and well-structured: it starts with a clear purpose, then explains the propose-then-confirm flow, includes usage examples, and is front-loaded with the most important info. No wasted words.

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?

Given the tool's complexity (one optional parameter, no output schema, but with safety considerations), the description is quite complete. It explains what happens, what it doesn't do, and how to use it. Minor gap: it doesn't describe what the output looks like (e.g., a list of accounts or a connection confirmation), but since there's no output schema, this would be helpful. However, the description still provides enough for an agent to use it effectively.

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 schema has one parameter, account_id, with no description in the schema. The tool description explains its use: 'call it again with account_id set to the one to connect,' which adds meaning. However, it doesn't specify the format of account_id (e.g., numeric string), but the schema coverage is 0%, so the description partially compensates but could be more detailed.

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 clearly states the tool's purpose: connecting an AWS account from the MCP client using existing local credentials. It distinguishes itself from siblings like connect_azure and connect_gcp by explicitly naming AWS and detailing its scope (no terminal needed, local-only).

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 description provides explicit usage guidance: call with no arguments first to list available accounts, then call again with account_id set. It also gives example prompts. While it doesn't explicitly mention alternatives, the context of siblings is clear, and the instructions are actionable.

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?

The description discloses non-obvious behavior: the tool does not accept or transmit the secret, it returns a Cloud Shell script, and the user must complete the connection in their own terminal with the secret encrypted into a local vault. This goes well beyond the generic annotations, and nothing contradicts readOnlyHint/destructiveHint.

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 description is front-loaded with the core purpose, then uses a focused rationale and examples. Each sentence contributes either the security constraint, the comparison to alternatives, or the concrete command to run; no filler.

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?

Given zero parameters and no output schema, the description fully covers what the tool returns (a Cloud Shell script), what the user must do next, why the tool exists, and example invocation phrasings. It is complete for safe 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?

The tool has zero parameters, and the description explains why: Azure credentials cannot be safely auto-detected, and a secret would have to be pasted into chat/tool args, which nable intentionally avoids. This adds crucial semantic meaning to the empty input 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?

Clearly states it is a guide for connecting Azure, with the specific security constraint of keeping the service-principal secret off the model. It distinguishes itself from connect_aws and connect_gcp by explaining their local-credential behavior vs Azure's need for a client secret.

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 contrasts with connect_aws and connect_gcp: those read machine-local credentials and require no sensitive pasting, while Azure requires a client secret that must not enter the chat. It also directs the user to finish with `finops setup azure` in their own terminal, making the usage path 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?

The description goes beyond annotations by disclosing that it never changes anything in GCP, keeps credentials locally, and operates in a propose-then-confirm manner. This adds crucial context about side effects, which aligns with the readOnlyHint:false annotation (since it may write local state) and clarifies it is not destructive.

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 description is well-structured with a clear summary, detailed explanation, usage steps, and examples. Every sentence serves a purpose, and the information is front-loaded—the first line conveys the core function, then expands on usage and safety.

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?

Given the tool's simplicity (one optional parameter, no output schema), the description is complete. It explains what happens with and without the parameter, reveals its safety profile (no GCP changes, local-only), and provides use-case examples, covering all necessary context for an agent to use the tool 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?

Even though the schema has 0% description coverage, the tool description fully explains the single optional parameter billing_account_id: it is the ID of the account to connect, and its absence triggers a listing behavior. This adds meaning beyond the bare 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?

The description clearly states the tool's purpose: to connect a Google Cloud billing account from within the MCP client. It uses a specific verb ('connect'), identifies the resource ('Google Cloud billing account'), and distinguishes itself from sibling tools like connect_aws and connect_azure by naming GCP 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 description provides explicit usage guidance: call with no arguments to list available billing accounts, then call again with billing_account_id to connect one. It also includes examples and explains the two-step flow, making it clear when to use this tool versus alternatives.

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 declare readOnlyHint: false while the description asserts 'Read-only: it estimates and checks, it never applies anything' — a direct contradiction. Per the rubric, any description that contradicts annotations scores 1 and must be flagged as an Annotation Contradiction.

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?

Well-structured and front-loaded: a one-line summary, read-only callout, bulleted input options, budget behavior, good triggers, and examples. Every sentence serves a purpose, and the bullet formatting makes it scannable.

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 complex 6-parameter, multi-mode tool with no output schema, the description covers inputs, verdicts (ok/warn/over_budget/no_budget), return values (delta, headroom), defaults, and edge cases. No significant gaps exist.

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?

Despite 0% schema description coverage, the description fully compensates by explaining each input mode (Terraform plan via json/file/dir, helm_diff, and the monthly_delta_usd escape hatch) and documenting budget_name's default (first active budget) and the no_budget edge case.

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 opening line 'Cost preflight for a proposed change: what it costs and whether it fits budget' uses a specific verb+resource construction, and the description further clarifies scope ('Read-only', 'it estimates and checks, it never applies anything'), distinguishing it from siblings like get_cost_summary and check_ai_budget.

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 excellent when-to-use guidance: 'Call this BEFORE applying an infrastructure change', a 'Good triggers' list, and two example queries. However, it lacks an explicit 'when not to use' note naming alternative tools (e.g., get_cost_summary), so it falls just short of a 5.

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=true and destructiveHint=false, so the safety profile is known. The description adds no extra behavioral context (e.g., time range defaults, granularity handling, or response shape). It doesn't contradict annotations, so a baseline 3 is appropriate.

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 description is extremely concise: one declarative sentence plus two illustrative examples. No wasted words, and the main functionality is front-loaded.

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 6 optional parameters, the description should explain return values and parameter semantics more thoroughly. It covers basic summarization but omits important details like filtering by provider, category, and date ranges, as well as default granularity behavior. This is a moderate-complexity tool left under-specified.

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% and there are 6 parameters, but the description only hints at groupings like service/account/region without mapping to actual parameter names (provider, account, category). The examples imply date-range usage but don't clarify date formats, granularity options, or null semantics. This is insufficient compensation for zero schema 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?

The description clearly states the tool's function: 'Get total spend summarized by service, account, and region.' This is a specific verb and resource with scope. It doesn't explicitly differentiate from siblings like estimate_change_cost, 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?

The examples ('How much did we spend last month?' and 'Give me an AWS cost summary for January') provide clear contextual usage for cost summary queries. There is no explicit exclusion or alternative comparison, which prevents a 5.

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=true and destructiveHint=false, so the safety profile is covered. The description adds useful context: it marks each provider as connected or not-configured, includes the active plan, and names the setup command for each missing provider. This goes beyond the schema and annotations, providing expected output details.

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 description is concise enough: two short paragraphs plus two example questions. The main functional description is front-loaded, and the examples clarify typical intents. Each sentence adds value—no filler or repetition of schema/annotations.

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 simple read-only list tool with no parameters and no output schema, the description fully covers what the agent needs to know: what it lists, how it labels providers, the active plan, and the setup command hint. It addresses common user questions (which providers, is GCP set up) and gives enough context for 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?

There are zero parameters and schema coverage is 100% (trivially). With no parameters to explain, the baseline is 4 as per rubric. The description doesn't need to add parameter meaning; it correctly focuses on behavior and output.

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 clearly states the tool lists all known providers with connection status and active plan, using specific verbs and resources. It distinguishes from sibling connect_* tools by focusing on listing status rather than establishing connections.

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 explicitly says this is the starting point for 'what am I connected to' and for spotting missing credentials, with setup commands named for each not-configured entry. While it doesn't explicitly name alternative tools, the sibling names (connect_aws, connect_azure) make it implicit that those are for setup. Lacks an explicit when-not-to-use clause.

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?

Despite readOnlyHint/destructiveHint annotations already covering safety, the description adds substantial behavioral context: detected ambient credentials, the one-command connect path, the prohibition on asking for secrets in chat, and when to direct users to their own terminal. This goes well beyond the annotations.

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 description is well-structured with a clear opening summary, usage triggers, hard rules, and examples. Each section earns its place, and the most critical information is front-loaded.

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?

Even without an output schema, the description tells the agent what the tool returns (connected providers, existing credentials, exact commands) and how to act on that information. It also includes examples, making it complete for a no-parameter status tool.

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 has zero parameters, so the baseline is 4. The description still explains what information the tool reports, which indirectly defines the empty input expectation, but there are no parameter semantics to clarify.

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 clearly states the tool's purpose: reporting onboarding status, existing credentials, and exact commands for remaining providers. It distinguishes itself from siblings like connect_aws and list_connected_providers by focusing on agent-driven setup status and the full onboarding picture.

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 says when to call ('when the user asks to connect a provider, says setup is incomplete, or asks what they are missing') and gives hard rules/preferences for how to handle credentials. It also names preferred zero-secret paths like `finops connect`, providing clear guidance over alternatives.

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=true and destructiveHint=false, so the description adds context beyond that: it mentions tailoring to connected services ('tailored to what you've connected') and explains the 'detailed' parameter's effect ('list the underlying tool names'). This provides useful behavioral detail without contradicting the annotations.

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 description is front-loaded with the primary purpose in the first sentence. It includes trigger phrases and examples, which are helpful but slightly redundant (e.g., multiple example queries). It remains appropriately sized for the tool's simplicity, staying under 150 words.

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 simple tool with a single optional boolean parameter and an existing output schema, the description is fully sufficient. It covers purpose, usage triggers, parameter behavior, and even situational guidance (post-connection). No critical information 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?

The schema has zero description coverage for the 'detailed' parameter. The description compensates by explicitly explaining its purpose: 'Pass detailed=True to also list the underlying tool names.' This adds meaningful semantic context beyond the schema's minimal definition.

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 clearly states 'Show everything nable can do, tailored to what you've connected.' This is a specific verb ('Show') and resource ('everything nable can do'). It distinguishes itself from sibling tools like connect_aws or get_cost_summary by focusing on overall capabilities, and the usage examples reinforce its purpose.

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 provides when-to-use triggers: 'Call this when the user asks "what can you do?"...' and lists multiple example queries. It also gives a strong directive: 'Always call it right after a user connects their first account' to showcase new capabilities. No alternatives are mentioned, but the guidance is clear and actionable.

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