data_series取「一個標的 × 一個欄位 × 一段時間」的時間序列，如 ETF 日收盤、基金淨值歷史。每點只有日期與值；資料缺口會以 gapNote 誠實標示、不靜默補值。Time series for one security and field.
| Name | Required | Description | Default |
|---|---|---|---|
| id | Yes | 標的識別碼，如 0050、T1104Y | |
| months | No | 回看月數，預設 12、上限 120 | |
| dataset | Yes | 資料集，如 etf.close／fund.nav（見 data_catalog） |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the transparency burden. It discloses return shape ('每點只有日期與值') and truthful gap handling ('資料缺口會以 gapNote 誠實標示、不靜默補值'), which is meaningful behavioral information beyond the schema. It omits ordering/pagination/error details, so not a 5.
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 compact and front-loaded, with the core usage pattern in the first clause and key output behavior in the second. The trailing English sentence 'Time series for one security and field' is slightly redundant with the Chinese opening, but the overall size is still appropriate.
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 3-parameter tool with no output schema and no annotations, the description covers enough: parameters are fully described in the schema, response shape (date/value) and gapNote behavior are stated, and dataset values are linked to data_catalog. The main unresolved context is how this generic series relates to siblings like etf_price_history or fund_nav_history.
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 schema already documents id, months, and dataset, including defaults and limits. The description's '一個標的 × 一個欄位 × 一段時間' maps conceptually to the parameters but adds little parameter-level detail beyond what the schema states.
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?
Description opens with a precise operation: 取「一個標的 × 一個欄位 × 一段時間」的時間序列, naming the action, resource scope, and time dimension. Examples (ETF 日收盤、基金淨值歷史) and '每點只有日期與值' make it easy to distinguish from single-point data_point or multi-column data_table siblings.
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?
It clearly defines when to use it: any need for a one-security, one-field historical series, with concrete use cases. It does not explicitly name alternatives or state when not to use data_table/data_query/etf_price_history, so it lacks exclusions but still gives strong context.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Most tools have clear boundaries, but there is notable overlap between data_series and etf_price_history/fund_nav_history, since all three can provide time-series data. market_sentiment and taiwan_market_overview also both expose Taiwan sentiment, creating potential selection ambiguity.
The data_* tools follow a clear prefix pattern, but the rest mix noun-style names (fx_rates, market_sentiment), object-action names (etf_lookup, fund_lookup), and generic verbs (search, fetch). The names are readable and understandable, but the overall convention is inconsistent.
At 15 tools, the count is at the upper edge of the ideal range and mostly reasonable for a Taiwan finance data server. However, several domain-specific wrappers duplicate capabilities already available through the generic data_* tools, so the set feels slightly heavier than necessary.
The generic data catalog plus data_point/data_query/data_series/data_table provides broad coverage for read-only financial data, and the domain tools cover ETFs, funds, FX, market overview, sentiment, and climate indicators. Minor gaps remain, such as no explicit Taiwan stock-specific lookup or price history tool, but the search and generic data layers partially compensate.