run_experiment

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

Annotations already mark the tool as destructive and non-idempotent, and the description adds meaningful context beyond that: the plan is 'single-use', meaning executing it consumes the plan, and the tool runs 'all current trials' rather than a subset. Directing the user to inspect results via get_experiment also gives useful postcondition context. No contradiction with annotations is present.

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 very short and front-loads the primary action, which is good for conciseness. The phrase 'then inspect get_experiment' is telegraphic and could be read as instructing the agent to inspect the tool itself rather than the experiment state via that tool, so the structure sacrifices clarity.

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 one required parameter, an output schema, and annotations covering destructive/non-idempotent behavior, the description does not need to explain return values or safety. Still, it omits important context such as how plan_token is created, what 'current trials' means, and whether execution is synchronous or asynchronous, so an agent has to infer several operational details.

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 schema has one parameter, plan_token, with 0% description coverage, so the description must compensate. It implicitly ties plan_token to 'one single-use plan', which gives some semantic meaning, and the schema provides a 64-character hex pattern. However, the description never explicitly explains that plan_token is the identifier returned by a planning step or how to obtain it, leaving a partial gap.

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 uses an explicit verb ('Execute') and names the resource ('all current trials from one single-use plan'), so the core action is clear. It also points to get_experiment as the follow-up inspection step, which helps distinguish it from planning tools like plan_experiment. However, 'current trials' is somewhat vague and 'inspect get_experiment' is grammatically ambiguous.

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 implies the tool should be used when a single-use plan exists and all of its trials need to be executed, and it suggests get_experiment as the next step. It does not explicitly state when not to use this tool or name alternatives such as run_fault_experiment or run_inference_profile, so the guidance is mainly implicit.

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