scigan

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

Even though readOnlyHint=true and openWorldHint=false are already in annotations, the description adds substantial behavioral context beyond that: the adversarial training is unstable across seeds, torch is imported lazily, and the generator recovers the counterfactual dose distribution. These details inform the agent about runtime behavior and variability that annotations cannot convey.

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 dense and well-organized, covering purpose, assumptions, pre-conditions, failure modes, alternatives, and minimum N in a compact block. Every sentence adds useful information, though it is longer than the minimal ideal and could benefit from clearer labeling (e.g., bullet points), which keeps it from a 5.

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 complex 11-parameter estimation tool, the description covers the essential decision factors: statistical assumptions, data requirements, software dependency, sample size floor, failure modes, and alternatives. An output schema exists, so return-value details are already structured. Nothing critical is missing for an agent to decide whether and how to invoke this tool.

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 coverage is 100%, so the baseline is 3. The description adds value by clarifying that the treatment must be continuous (a dose), which is not stated in the schema's treatment property ('Treatment indicator, treatment variable, or treatment array'). This extra constraint helps the agent verify inputs before calling.

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 states the tool is an 'Adversarial dose-response estimator (Bica et al. 2020)', which clearly conveys that it estimates dose-response curves using a specific adversarial method. It lacks an explicit action verb like 'estimates', but the noun phrase is functionally clear. It also distinguishes itself from sp.vcnet and sp.dose_response by naming them as alternatives.

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 provides explicit when-to-use guidance: it lists assumptions, pre-conditions (continuous treatment, torch installed), and typical minimum N. It gives clear failure-mode guidance—'dose-response estimates vary across seeds'—and tells the agent to average across seeds or use an alternative. This is strong, actionable routing information.

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