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

Annotations only declare readOnlyHint and openWorldHint; the description adds substantial computation semantics beyond that — counting starts only after unit assignment, denominators are enrolled units over the experiment's run, non-control groups are compared with a two-sided test yielding lift, 95% CI, p-value and one of four verdicts, plus a sample-ratio check. That is exactly the kind of behavior an agent needs to trust the output.

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 purpose, then prerequisites, then result shape; no sentence is filler. It is one dense block though — the metric-kind enumeration would scan better as a list, which is the only structural weakness.

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, the description takes on the burden of describing results and does so completely (lift, CI, p-value, verdict values, sample-ratio check), and it covers result-window semantics for the range parameter. Nothing an agent needs to call this 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?

The hardest parameter, `metrics`, is fully disambiguated: all three `kind` variants are defined, and `sum_per_unit`'s `value` (argument vs property) is explained with a concrete example ('revenue per player'). The remaining parameters (range, projectId, environment, controlGroupId) already carry schema descriptions, so the description's silence there is acceptable at 67% 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?

Starts with a specific verb and resource: 'Measure how an experiment's groups perform on up to 5 metrics, with statistics.' The scope (per-group, up to 5 metrics, statistical comparison vs control) is precise enough to distinguish it from get_experiment or preview_experiment without opening any 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 prerequisites: resolve event/argument names via list_event_types / get_property_values first, and load the understand-experiments skill before calling. It does not, however, contrast this tool with named siblings such as preview_experiment or execute_query, so the when-not path is left implicit.

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