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

Annotations already cover the safety profile (readOnlyHint=true, destructiveHint=false, openWorldHint=false), so the description only needs to add context beyond that. It contributes the return shape ('PSI 值与稳定性判断'), which is genuinely useful given there is no output schema, but it says nothing about default binning, whether the judgment uses standard 0.1/0.25 thresholds, or any input constraints (equal-length arrays, probability sums).

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 pipe-delimited, front-loaded format puts the purpose first and quarantines aliases and return info into labeled segments, so nothing critical is buried. The alias list (PSI, 模型稳定性, 群体稳定性, population stability, 漂移) is somewhat padded for a tool-name-based selector, but it does not obstruct scanning.

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 simple read-only, closed-world computation with a fully documented 3-parameter schema, the definition covers the essentials: what it computes, when to reach for it, and roughly what comes back. What is missing is interpretive guidance an agent would need to report results correctly — the meaning of the returned stability verdict and its thresholds — which matters because no output schema exists to explain it.

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% — all three parameters carry inline Chinese descriptions (分箱数, 开发集概率, 验证集概率) — so the schema itself carries the parameter meaning. The description adds no additional semantics such as array-length matching or a default for n_bins, which is the expected baseline-3 outcome when the schema already documents everything.

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 definition names a specific statistical quantity (population stability index) and states the two inputs it compares (expected vs observed score distributions), so an agent knows exactly what it computes. It is clearly distinct from every sibling in the list (no other stability/PSI tool exists). It stops short of a full 5 because the verb is only implied via the 'USE WHEN:' clause rather than stated outright.

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 'USE WHEN:' clause gives a concrete triggering condition — comparing expected versus observed score distributions — which is enough for an agent to know when this tool applies. It offers no exclusions or named alternative (e.g. it does not say to compare against roc_auc or calibration_curve for discrimination/calibration rather than stability), so it falls short of the 5 tier.

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