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

With no annotations, the description carries the full burden and does: comma-separated codes are accepted but return flat rows with no per-symbol status and invalid codes are silently dropped ('sh600519,sz999999 returns only 1 row with no hint'), HK and sector codes work but return missing fields, ETF calls only populate code/name/listedDate, and results are a static 24-hour cache. This is exactly the failure-mode context an agent needs.

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 the field list and trigger questions, then structured with ⚠️ markers so each caveat is scannable. It is on the long side and the multi-code warning is partly restated in the parameter schema, but almost every sentence prevents a concrete misuse.

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 one-parameter read tool with an output schema, everything an agent needs is present: what is returned, what is citable, which codes are valid, which assets are out of scope, where the alternative tools are, and how caching affects repeated calls. Nothing material is left to inference.

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% (baseline 3), but the description adds real meaning beyond it: 'sh600667' or bare '600667' both work because the prefix is auto-appended, and although the schema says multi-code goes to batch, the description clarifies what actually happens if you pass commas anyway (flattened rows, silent drops). That is genuine value over the schema text.

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?

Names a specific resource (company profile: listing date, main business, industry, website, registered capital, chairman, headcount) and states in one line the user questions it answers. It also explicitly differentiates from the nearest sibling, ashare_snapshot: 'snapshot gives valuation/market cap/share count (numeric side), this tool gives who the company is (business side)'.

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 explicit when-to-use phrasing ('what does this company do / when was it listed / what industry / who is the chairman'), and routes alternatives: valuation questions → ashare_snapshot, multiple codes → ashare_profile_batch, ETFs → ashare_etf(kind='详情'). It also states when-not-to-use: ETFs and indices are excluded because business/industry/sector are absent at the source.

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