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

With no annotations, the description carries the burden and does well: it defines the conversion counting rule (only at/after first exposure), states the default ordering (running first), the cost ('Free'), and a critical interpretation constraint. It omits read-only confirmation, auth requirements, and result volume/pagination limits.

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?

Highly dense and front-loaded: the payload contents come first, then three terse constraint sentences. Every clause carries information (metrics, statistical method, verdict options) with 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?

With no output schema, the description usefully enumerates the return shape and verdict vocabulary, and covers the optional-key behavior. It is only incomplete because 'site' is undocumented and no volume/pagination expectation is set for the workspace-wide case.

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 optional 'key' param well (omitting it returns every test in the workspace, running first) but says nothing about 'site', leaving one of two parameters entirely uninterpreted.

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 enumerates exactly what the tool returns (exposed/converted counts per variant, rates with 95% intervals, chance to beat control, lift, sample ratio check, verdict), which unambiguously identifies it as the read path for an A/B test. Combined with the title 'Read an A/B test' and siblings experiment_create/experiment_close, an agent can tell it apart without opening a 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?

It gives a scoping rule ('Without a key, every test in the workspace, running first') and an interpretive rule ('Never call a winner the verdict does not'), which implies usage but never states when to call this versus experiment_close or any alternative. No explicit when-not guidance, so usage is only implied.

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