get_taiwan_law_versions查立法院開放資料平台（官方直連）取得法律的三讀通過紀錄與附帶決議：三讀屆會次、三讀通過條文 PDF、附帶決議內容。涵蓋近年（2023 起）三讀紀錄；完整修正沿革請併用 search_taiwan_laws 查全國法規資料庫。資料授權：政府資料開放授權條款第 1 版。
| Name | Required | Description | Default |
|---|---|---|---|
| law | Yes | 法律名稱（如：道路交通管理處罰條例） | |
| limit | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description discloses the data source (official Legislative Yuan platform), the time range (2023 onwards), the nature of output (third-reading PDFs and resolutions), and the data license. It does not discuss rate limits or authentication, but for a read-only query tool, this is adequate behavioral context.
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 three sentences, front-loaded with the primary action, and includes only relevant information (purpose, coverage/alternative, license). No redundant or tautological content.
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?
The tool has no output schema or annotations, but the description explains the return content and the coverage limitation. It does not mention error handling or response structure, but given the simple two-parameter tool, it is reasonably complete.
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?
The description does not explain the parameters. The schema covers the law parameter but not limit, and the description adds no details about the limit parameter or how results are paginated. With 50% schema coverage, this is a gap.
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 description clearly states the tool queries the Legislative Yuan Open Data Platform for third-reading passage records and attached resolutions, listing specific outputs (session number, PDF, resolutions). It also differentiates from the sibling search_taiwan_laws by noting that tool covers full amendment history.
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 explicitly states that for complete amendment history, users should also use search_taiwan_laws, and clarifies the coverage begins in 2023. This tells the agent when to prefer this tool (recent third-reading data) and what alternative to use for full history.
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.
Each tool targets a distinct legal source or action: laws, regulations, interpretations, judgments, MOJ documents, bills, legislative history, and intent analysis. Even similar tools like search_taiwan_laws vs search_taiwan_regulations are clearly separated by scope (full-text vs name-only) and descriptions provide sufficient boundary definitions.
All tool names follow a consistent verb_noun pattern with a domain prefix (taiwan or moj). The verbs are limited to get, search, and analyze, and the object nouns are unambiguous (e.g., interpretations, judgments, laws, regulations, bills). No mixed conventions or vague verbs.
14 tools is well-scoped for a comprehensive Taiwan legal research server, covering search, retrieval, citation extraction, intent routing, and legislative tracking. Each tool has a clear purpose and the count is within the ideal 3-15 range.
The tool surface covers the full lifecycle of legal research: finding and retrieving laws, regulations, interpretations, judgments, and bills; obtaining legislative history and versions; converting law names to PCODEs; analyzing user intent; and even extracting citations. No obvious gaps that would block an agent's legal research workflow.