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

Annotations already declare readOnlyHint, idempotentHint and destructiveHint=false, so the safety profile is covered; the description's "nothing is written and nothing is charged" largely restates that. What it does add beyond the annotations is that the returned figure is an upper bound ("the most it may cost"), which prevents an agent from treating it as an exact quote, but it says nothing about pricing fluctuations, latency, or failure modes for invalid input combinations.

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?

Two short sentences, front-loaded with the purpose and cost semantics, then the input rule. No filler and nothing repeated from the schema text.

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 no-side-effect estimator with no output schema, it covers purpose, return semantics, and invocation shape adequately. A small gap remains on what happens when inputs are omitted or mismatched with the selected tool/model, and whether the estimate can be passed to start_run.

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?

With 100% schema coverage the baseline is 3, and the description earns more by stating "Give one of tool, model and workflow, and the inputs" – expressing mutual exclusivity among the three selector parameters that the flat schema cannot convey and that matters given zero required parameters. It remains slightly ambiguous about whether exactly one selector plus inputs is required or whether selectors are optional.

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 ("Prices"), the resource ("a run"), and the crucial scope modifier "without starting it," plus what is returned ("the most it may cost, in credits"). This clearly separates it from the sibling start_run without opening either 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?

The phrase "without starting it" plus "nothing is charged" paints a clear pre-flight / dry-run context that implies estimate before calling start_run. It stops short of explicitly naming start_run as the alternative or stating when estimation is unnecessary, so it is clear context without explicit exclusions.

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