ips

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

Annotations already declare readOnlyHint=true and openWorldHint=false; the description adds non-obvious behavior: positivity assumptions, no unmeasured confounding, high-variance failure from extreme importance weights, and a typical minimum N. It does not discuss alpha/confidence-interval behavior, but the output schema exists to carry return-shape information.

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 but well-structured with labeled sections: Assumptions, Pre-conditions, Failure modes, Alternatives, and Typical minimum N. Every clause adds information; slight redundancy like repeated 'logging policy' phrasing prevents a 5, but there is no filler.

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 13-parameter OPE tool with an output schema and read-only annotation, the description supplies the key invocation context: required inputs, assumptions, failure modes, how to fall back to alternative estimators, and even a minimum sample-size heuristic. It is complete enough for an agent to invoke correctly without additional round-trips.

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, but the description adds real parameter meaning by clarifying X, A, R as context, logged action and reward, and pointing to logging-policy propensities. It does not elaborate on clip, alpha, detail, or as_handle, though those are already described in the schema.

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 opens with 'Inverse propensity score OPE,' naming the estimator family and task precisely. It then refers to X/A/R and explicitly lists sibling alternatives like sp.snips, sp.doubly_robust, and sp.direct_method, so an agent can distinguish this tool from closely related tools.

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 failure-modes note gives a concrete rule: when the target policy diverges from logging, prefer snips or doubly-robust to reduce variance, and the Alternatives line names direct_method. It does not give a crisp positive 'use IPS when...' trigger, but the assumptions and preconditions imply the appropriate context.

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