Experiment Rollouts

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

The readOnlyHint annotation already establishes that this is a safe read. The description usefully adds that results are 'persisted' (stored, not computed on the fly) and 'bounded' (paginated rather than unbounded), which goes beyond the annotation. It stops short of describing page-size limits, ordering, or consistency behavior.

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 tight sentence with no filler and the key scoping concept ('bounded page') front-loaded. It is efficient, though arguably too sparse for a tool with three undocumented parameters.

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?

An output schema is present, so return values need no explanation, and the annotation covers the safety profile. But with zero parameter documentation and no usage context, the definition is insufficient for an agent to know what run_id means or how paging works.

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 all three parameters (run_id, limit, offset) are undocumented anywhere. The description's phrase 'bounded page' loosely implies limit/offset semantics but never explains them, and the required run_id is not described at all, leaving the agent to guess what identifier it must supply.

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 verb 'Read' plus the resource 'persisted rollout or evaluation trajectories' is specific and concrete, and 'bounded page' signals pagination. It does not, however, name or distinguish itself from nearby siblings like evaluation_get, training_metrics, or training_logs, which an agent would need to route correctly.

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?

There is no when-to-use guidance, no prerequisites, and no mention of alternative tools for retrieving trajectories or metrics. The agent is left to infer that this is the tool for paging through rollout data rather than any of the ~45 sibling tools.

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