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

Annotations already mark this as a non-read-only, non-destructive, non-idempotent operation, so the safety profile is covered. The description adds value beyond that: it discloses cost (spends the workspace's AI budget), an authorization requirement (Admins only), and a persistence side effect (the run is saved in Training → Evals). It does not describe runtime behavior such as whether grading is synchronous or how long the run takes.

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?

A single sentence packed with the highest-value facts, ordered action → outcome → cost → permission. Zero filler and nothing repeated from the title.

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 mutation tool with no output schema and full schema coverage, the description supplies the non-obvious operational facts an agent needs (cost, admin gate, persisted result location). It is still silent on result shape and timing, which is a minor gap given no output schema exists.

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 100%, so the baseline is 3; the description still adds meaning by framing `cases` as test questions and the `expect` field as the grading target, and by surfacing the 20-question cap in prose rather than leaving it buried in maxItems. It adds no guidance on `model` beyond what the schema already says.

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?

States a specific verb-action chain (ask the product's AI agent test questions and grade the answers) plus the resource (product agent) and a concrete scope limit (up to 20 questions). This cleanly separates it from look-alike siblings such as test_journey, test_proactive_message, and get_training_overview.

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 real operating context for deciding whether to invoke: it consumes the workspace's AI budget and is restricted to admins, and it says the run lands in Training → Evals. It stops short of naming alternatives or a when-not-to-use condition (e.g. lightweight single-question testing), so it is clear context without exclusions.

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