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

With no annotations provided, the description carries the full burden of behavioral disclosure. It reveals that the tool only extracts citations within the reasoning, traces backward to earlier citations, and optionally includes surrounding text. This goes beyond the schema by specifying the 80-character context length and temporal direction. It doesn't detail edge cases (e.g., no citations found, invalid case_id), but for a read-only extraction tool, the key behaviors are sufficiently disclosed.

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 a single, concise sentence that leads with the action ('從...理由書中抽取') and front-loads the core functionality. Every element earns its place—describing the source, the extraction type, the backward-citation direction, and the optional context. No wasted words.

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?

Given the tool's simplicity (2 parameters, no output schema), the description is complete enough to understand the return value: a list of cited interpretation numbers with optional context. It covers the input generation and output type without needing an explicit output schema. Minor gaps include unspecified response structure when no citations exist, but the core usage is fully specified.

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%, so the baseline is 3. The description adds value by elaborating on include_context: it specifies that the excerpt is '前後 80 字片段' (80 characters before and after), which is more specific than the schema's '每個引用附上原文片段'. It also reinforces case_id's format by referencing get_taiwan_interpretation, aiding correct invocation.

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 extracts cited interpretation numbers from the reasoning of a specified 釋字/憲判字, with an optional 80-character context excerpt. This specific verb-resource pairing ('抽取引用的其他釋字與憲判字字號') distinguishes it from sibling tools like get_taiwan_interpretation (retrieving a single interpretation) and search_taiwan_interpretations (searching by keywords).

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 description conveys a clear context: given a case_id, retrieve the older interpretations it cites, going backward in time. It implies the tool is used for citation relationships rather than general retrieval, but it does not explicitly name alternatives or state when not to use it. The '往前追溯' phrasing provides directional context, making the use case clear enough.

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