overlap_weights

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

Annotations already declare readOnlyHint=true and openWorldHint=false; the description adds meaningful behavioral context by disclosing failure modes (extreme propensity scores, residual imbalance) and remediation steps, as well as a typical minimum N of 200. It does not mention the caching side effect of as_handle=true, but that is parameterized behavior is absent from the description, so it doesn't contradict the read-only annotation.

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 substantial but organized with labeled sections (Validation, Assumptions, Pre-conditions, Failure modes, Alternatives, Typical minimum N) and front-loads the core estimator identity. Each sentence earns its place, though 'certified parity evidence' is ambiguous and slightly cryptic.

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 14-parameter estimator with an output schema, the description covers statistical context (assumptions, pre-conditions, failure modes, alternatives, minimum N) while the schema fully documents parameters and output. Nothing critical for tool invocation is missing; only the 'Validation: certified parity evidence' line is opaque.

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 100%, so the baseline holds. The description adds contextual rationale for trim (rarely needed, clip to [trim, 1-trim]) and implicitly for estimand via the ATO default, but it does not systematically describe each of the 14 parameters; the schema carries that burden.

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 a specific method-object pair, 'Overlap-weight (ATO) treatment effect estimator,' which clearly identifies the estimator and its target. It further differentiates from siblings by explicitly listing alternatives (sp.propensity_score, sp.cbps, sp.ebalance, sp.dml), so an agent can select it without inspecting the 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 description provides explicit assumptions (unconfoundedness, overlap, back-door closure) and pre-conditions (pre-treatment covariates, binary treatment, sufficient overlap), which collectively signal when the tool is appropriate. Failure modes also advise remedies like trimming or re-specifying with CBPS/entropy balancing. However, it doesn't directly state when to prefer ATO over ATE/ATT/ATC as an estimand choice beyond the default.

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