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

Annotations already declare readOnly/openWorld/idempotent/non-destructive, so the safety bar is low. The description adds genuinely useful behavioral context beyond that: default concept universe (kpl) with the auto-fallback semantics, and a hard output requirement that every stock be rendered as a markdown quote link. It does not describe pagination or result-size behavior, keeping it out of 5 territory.

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?

Front-loaded with purpose and capability list, which is good, but it is dense and redundant — the instruction that the agent should map natural-language conditions to structured params is stated twice (once in the body, again under 何时使用) and the worked examples overlap the schema examples. Sentences earn their place individually, but the block as a whole is longer than needed.

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 64-parameter screener with an output schema and full annotation coverage, the description covers what the agent actually needs: purpose, NL→param mapping strategy, concept-source defaults/fallback, and output link formatting. It omits defaulting behavior for date/limit and any explicit 'no results' handling, but these are secondary given the rich schema.

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% and the schema itself already carries per-field mapping examples (e.g. aboveMa, highPctChgMin, marketCapMinYi), so the baseline is 3. The description's cross-cutting sentence-to-params examples are mildly additive but largely duplicate what the schema descriptions already say, so no uplift is warranted.

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?

States a specific verb+resource ("主条件选股工具") and enumerates the full filter surface (概念/题材、行业、地区、价格、市值、涨跌幅、均线、放量、板块、ST 状态等). It also positions itself against alternatives by telling the agent not to hand-assemble concept_stocks+kline+valuation_snapshot. An agent can identify this as the general-purpose screener without opening the 64-param 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?

Gives concrete routing guidance: map 问财-style natural language directly to schema params, with worked examples ("半导体板块、五日线之上、放量三倍、上海、名字带龙" → named params) and an explicit anti-pattern (don't stitch together concept_stocks+kline+valuation_snapshot). The "何时使用" section is partly circular, and no explicit when-not-to-use versus siblings like limit_up_filter is given, so it stops short of a 5.

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