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

Annotations already mark this read-only and non-destructive, and the description adds substantial behavioral context beyond that: filterCel omitted means unfiltered rather than AWS-only, nestingEdges warns against summing a contributor with its ancestors/descendants, and aggregationMethod should usually be omitted. These are exactly the kind of non-obvious behaviors an agent needs to know.

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 information-dense and front-loaded with the core purpose, followed by usage rules and illustrative examples. There is some redundancy with the schema (e.g., compare {} auto-derivation and aggregationMethod guidance), but the examples and edge-case warnings justify the length.

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 8-parameter tool with nested objects and no output schema, the description covers the key invocation decisions: date windowing, compare behavior, column selection limits, filterCel semantics, nestingEdges math, and aggregationMethod. It does not describe the return format, but the examples and parameter clarifications make correct invocation achievable. Minor gaps remain around datePreset versus explicit from/to interaction.

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 high (88%), so the baseline is 3, but the description adds meaning through concrete examples and explains important cross-parameter behavior: compare can be {} for auto-derivation, filterCel's unfiltered semantics are clarified, columns can be strings or { column, contains } objects, and independent contributors may be summed. This goes beyond the raw 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 opens with a clear, specific statement: 'Find which dimension values drove a cost change between two periods.' It frames this as a before/after analysis, distinguishes it from related investigation tools by pointing to suggest_groupby / search / get_context as precursors, and provides concrete examples that remove ambiguity about what the tool does.

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 explicitly states when to use the tool: for before/after cost-change analysis, with compare required and {} auto-deriving the previous window. It also gives workflow guidance ('Prefer suggest_groupby / search / get_context first') and limits column selection to 2–4 dimensions (max 8), all of which helps an agent decide when and how to invoke it.

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