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

Annotations already declare readOnlyHint=true, destructiveHint=false and openWorldHint=false, so the safety profile is covered and the description needs only to add context. The 'RETURNS' line usefully discloses that per-covariate SMDs plus a balance decision come back, which matters without an output schema. However, it never states the balance threshold applied (e.g. 0.1) or how the judgement is derived, which is the one behavioral detail an agent would actually want.

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 piped metadata format front-loads the actionable 'USE WHEN' and 'RETURNS' fields, so the key facts are read first. The bilingual alias string is somewhat long but serves a real retrieval purpose for Chinese-language requests. No filler sentences, though the duplicated title is mild redundancy.

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 two-parameter, read-only computation the definition covers the essentials: what it computes, when to reach for it, and what comes back. The gaps are minor — no output schema exists, and the description does not explain the balance-decision criterion, but the RETURNS line keeps it usable as-is.

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%: 'treatment' is documented as group labels 0/1 and 'covariates' as a sample-by-variable matrix, so the schema carries the parameter burden. The description adds no further semantics — no note on ordering alignment between the group vector and matrix rows, nor on handling of categorical covariates. Baseline 3 is appropriate.

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 'USE WHEN' clause states the exact statistic and setting: standardized mean differences before/after PSM or IPTW for covariate balance. That is a specific analytic operation rather than a restatement of the name. It does not explicitly contrast itself with likely adjacent siblings such as propensity_score_matching or iptw_weighting, which supply the matched/weighted data this tool then checks.

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 gives a clear triggering context (post-PSM or post-IPTW balance assessment) and the ALIASES list helps an agent recognize synonymous requests like 'love plot' or '协变量平衡'. It stops short of the 5 bar because it names no exclusions or explicit alternatives — an agent must infer that raw unmatched comparisons or the upstream matching itself belong to other tools.

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