Server Quality Checklist

Tool Scores

• metric_shift_inspect_dataA
  4.4/5.0
  Behavior4/5

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

  Annotations already declare the tool as read-only, idempotent, and non-destructive, so the safety profile is well covered. The description adds valuable behavioral context by revealing that the tool performs datetime column estimation and returns dimension/metric candidates, which conveys that it does heuristic analysis rather than just raw listing. No contraindication with annotations.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is two compact sentences, front-loading the core purpose and then providing practical workflow guidance. Every sentence earns its place—no fluff, no repetition of schema details.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness4/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Given the tool's moderate complexity (3 optional params, no output schema, one sibling), the description covers the essential context: what it does, what it returns, and its place in the workflow. It could theoretically mention error handling or return format specifics, but the schema already handles param constraints, and the description sufficiently orients the agent for a read-only inspection tool.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters3/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema description coverage is 100%, with each parameter (rows, file_path, datetime_column) already well-documented including format constraints and mutual exclusivity. The description does not add substantive parameter-level detail beyond what the schema provides, so the baseline of 3 applies.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description uses a specific verb ('確認する' - inspect) and clearly identifies the resource (data contents). It enumerates the concrete outputs (columns, datetime column estimation, date range, dimension/metric candidates), which precisely scopes the tool's function. It also distinguishes itself from the sibling metric_shift_localize by explicitly positioning this as the pre-check step.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines5/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The description states exactly when to use the tool: before analysis and before calling metric_shift_localize. It explicitly recommends verifying column names and date range here prior to localization, giving clear workflow context. While it doesn't mention when not to use it, the recommendation is unambiguous.

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

• metric_shift_localizeA
  4.4/5.0
  Behavior4/5

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

  Annotations already declare readOnlyHint, idempotentHint, and destructiveHint, so the safety profile is known. The description adds valuable behavioral context: the day-of-week correction, the Squeeze algorithm, and the explicit statement that results are correlation-based and not causal proof. No contradiction with annotations.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is compact—four sentences in Japanese—and front-loaded with the primary purpose. Each sentence adds meaningful information: the question answered, the method (prediction + deviation), the algorithm/limitation, and a practical prerequisite. No unnecessary filler.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness4/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Given the tool's complexity (13 parameters, nested objects, no output schema), the description adequately covers the core logic and usage context. It explains what the tool does and how to interpret results, though it does not describe the output format in detail. The prerequisite hint about the sibling tool rounds out the picture.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters4/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema description coverage is 100%, so each parameter's meaning is well-documented in the schema. The description adds extra semantic context by giving an example attribute combination (weekday=Tue AND timeband=evening) and by clarifying the role of dimensions/reserved words, which helps the agent map the analysis concept to parameters.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description starts with a concrete question ('Where did sales drop?') and clearly states the tool's function: build a day-of-week-adjusted prediction from the baseline and identify attribute combinations where the deviation concentrates. It explicitly names the algorithm (Squeeze) and distinguishes itself from the sibling tool by directing users to metric_shift_inspect_data when column names are unknown.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines4/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The description clearly frames the usage context—localizing where a metric change occurred. It explicitly mentions using metric_shift_inspect_data first if column names are unknown, which is a concrete alternative. It also includes a caution about correlation vs. causation, but does not enumerate additional when-not-to-use scenarios.

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