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

With no annotations, the description carries the full burden. It usefully discloses the extraction source (official HTML or PDF) and a fallback behavior (scanned files that cannot be extracted return a PDF link). It does not cover permissions, rate limits, or other operational traits, so the disclosure is partial.

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 purpose, followed by a parenthetical on source/fallback, then a compact Args section. The enumeration of document types is slightly long but informative rather than wasteful.

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?

No output schema and no annotations, yet the description conveys what is returned (full text) and the edge-case return (PDF link for unextractable scans). Combined with the parameter provenance, this is nearly complete for a single-parameter fetch tool, with only auth/rate context missing.

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 0% and the single parameter is only titled 'Decision Id' in the schema. The description compensates well: it explains the parameter is the id returned by search_administrative_decisions and gives a concrete format example ('ey:A-115-000633'), adding semantics the schema lacks.

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 (取得/retrieve) and resource (the full text of administrative decisions), and enumerates the document types covered (appeal decisions, dispositions, review judgments, rulings, penalty cases, Control Yuan texts, attorney disciplinary resolutions). It is clearly the retrieval counterpart to search_administrative_decisions, though it does not explicitly distinguish itself from the many other 'get_*' 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?

Usage is implied rather than stated: by tying decision_id to 'search_administrative_decisions 回傳的 id', it signals the search-then-fetch workflow. However, there is no explicit when-to-use/when-not guidance or comparison against sibling retrieval tools like get_judgment or get_precedent.

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

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

With no annotations, the description carries the behavioral burden and does disclose notable traits: it states that recipient lists are omitted from the returned full text, and it explains the status/status_note semantics including what an absent status means. It does not cover error/not-found behavior, which is a minor gap for a read-only getter.

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-loads the purpose in one sentence, then uses clear Args/Returns blocks. The Returns list is long but every field name is informative; nothing is redundant boilerplate.

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?

There is no output schema, and the description compensates by enumerating the return fields (agency, doc_number, date, summary, full_text, related_laws, notes, status, attachments, source_url). Combined with the parameter source and omission notes, an agent has enough to call and interpret the result, though error handling is unaddressed.

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 0%, so the description must compensate, and it does: it identifies the single parameter's origin (search_agency_interpretations output) and gives two concrete example id formats (moj:FE393340, mol:e:勞動條 3:1100130312). This is far more than the bare "string" schema provides.

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 (取得) and resource (單一行政函釋全文) and scopes it to a single document, which clearly distinguishes it from the sibling search_agency_interpretations. It also discloses what is deliberately omitted (正本、副本受文者清單), so an agent knows the exact nature of the returned text.

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 Args section explicitly ties the call to search_agency_interpretations by telling the agent the id comes from that tool's return value, which effectively documents the search-then-fetch workflow. It stops short of stating any when-not-to-use conditions or naming other alternatives such as get_interpretation.

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

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

No annotations are provided, so the description carries the full behavioral burden. It does disclose the two operating modes and what include_context does (80-character surrounding snippet), which is real behavioral value, but says nothing about return structure, pagination, or error/empty cases.

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 front-loaded with the resource, then the two direction modes, then the alternative tool, then the args. Each sentence earns its place and the layout is easy to scan, though the arg block partially restates schema entries.

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?

With no annotations and no output schema and 3 parameters, the description covers the necessary ground: what it returns in each direction, all parameter meanings, and the routing to a sibling. Only finer detail about return formatting is absent, which the directions largely imply.

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 0%, so the description must compensate, and it documents all three params: case_id (format matching get_interpretation), include_context (attaches ~80 chars of source text), and direction with both allowed values explained. This adds clear meaning beyond the bare schema, though case_id format is only referenced indirectly via another tool.

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 names the specific resource (citation relationships between 大法官解釋/憲判字) and the specific operation for each direction, clearly stating the verb+resource. It explicitly distinguishes itself from the sibling search_judgments and names it, letting an agent tell the two apart without opening a 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?

It states precisely when to use each mode: direction="cites" for backward tracing from the reasoning text, direction="cited_by" for forward tracing from later texts. It also names the alternative (search_judgments) with the exact condition and keyword format for choosing it, leaving nothing to inference.

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

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

With no annotations, the description carries the full burden and does disclose meaningful traits: keyword matching runs only on unpunctuated official text and is limited to 憲判字 and pending cases, early 釋字 lack PDFs with full petition text in petition_text, and scanned PDFs may contain OCR errors. It does not cover auth requirements or rate limits, but the data-quality and scope caveats are the salient ones for a retrieval tool.

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 dense but well organized, leading with the resource inventory before the numbered usage modes and arg explanations. Every sentence carries information, though the sheer length and tight CJK formatting make it heavier than strictly necessary.

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?

With no output schema and no annotations, the description still details what each call returns (id, group, title, url; announcements; case fields such as original case number, consolidation, petitioner). An agent has everything needed to call it correctly across all three modes.

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 0%, so the description must compensate entirely, and it does: each of the three parameters is explained with accepted formats (case number styles, docket:/hearing:/amicus: ids, numeric vs news: document ids), example keywords, and the behavioral consequence of each combination.

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 names a specific verb+resource (retrieving Constitutional Court case-file documents) and enumerates the document types covered. It explicitly distinguishes itself from the sibling get_interpretation, noting that judgment text and full opinions live there instead.

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?

Three numbered modes state exactly which argument combinations to pass and what each returns (case_id alone, case_id+keyword, document_id alone). It also names the alternative tool (get_interpretation) and the condition that routes to it, leaving nothing to inference.

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

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

With no annotations, the description carries the full burden and does substantial work: it discloses cache-only default behavior (no network), the two retrieval modes, keyword override precedence (reasoning_keyword covers include_reasoning), and even names a returned field (opinion_documents with titles, court PDF links, word counts). It omits error/not-found behavior for invalid case ids, which keeps it short of 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?

Front-loaded with purpose, then case_id formats, then per-argument notes; each line earns its place. Slight redundancy from restating parameter defaults that also appear in the schema, and the bilingual framing adds length, but nothing is wasted.

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 7-param tool with no annotations, no output schema, and 0% schema coverage, the description supplies modes, caching behavior, argument semantics, and a return-field hint. The main gap is failure handling for unrecognized case ids, but overall an agent has enough to invoke it correctly.

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 0%, so the description must compensate, and it does: every one of the seven params is documented in the Args block, including precedence rules ('reasoning_keyword overrides include_reasoning'), filtering semantics for opinion_document, and the default/purpose of opinions_offset. The case_id format examples (釋字第748號 / 釋字748 / 748 / 111年憲判字第1號) are especially valuable given the empty schema.

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 (取得/fetch full text) and precise resource (大法官解釋 釋字 1-813號 / 憲法法庭裁判 憲判字), with an explicit coverage range. It is clearly the retrieval counterpart to the sibling search_interpretations, so an agent can distinguish it without opening the 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?

Explains the layered behavior (default layer served instantly from local cache with no network) and when to use the reasoning/opinion modes versus keyword snippet modes. It does not explicitly rate itself against search_interpretations or get_constitutional_case_file, but the context for choosing modes is clear.

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

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

With no annotations, the description carries the full burden and does well: it discloses the JID-over-URL priority, the rich return structure, and a genuine operational caveat (inspect history to check whether the judgment was overturned or remanded before citing). It omits auth/rate-limit or failure-mode behavior, so it stops short of 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?

Front-loads the purpose, then structures the two query methods and the return fields with an explicit caveat. Slightly long but every section (args, returns, history warning) carries load, so size is justified by the tool's complexity.

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?

With no output schema and no annotations, the description compensates by enumerating the return fields (case_id, main_text, facts, reasoning, cited_statutes, history, etc.) and explaining history_note. Complete enough to invoke correctly, though the absence of the input schema descriptions means ambiguous dual-input behavior goes unaddressed.

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 0%, yet the description documents both parameters with concrete format examples for jid (TPSV,104,台上,472,20150326,1) and url, and explains that jid is preferred. It does not clarify what happens if both or neither are supplied, keeping it below 5.

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 (取得單一裁判書全文 – retrieve a single judgment's full text) and explicitly scopes to a single document, distinguishing it from the sibling search_judgments. An agent can tell immediately this is the full-text fetch step after a search.

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?

Explains the two lookup paths (by JID via Open Data API, by URL via page load) and states the priority order. It also notes the JID comes from search results, implying the search→get workflow. It does not, however, explicitly name search_judgments as the alternative or state when not to use this tool.

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

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

With no annotations, the description carries the full burden and does add real behavioral context: full text is returned only when publicly available, and NCL-licensed full text is restricted to personal reading with no redistribution. It still omits return format and error behavior.

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?

Three short, front-loaded blocks: what it returns, the licensing caveat, then the argument. Every part earns its place, though the licensing sentence is a legal notice rather than invocation guidance.

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 single-param getter with no output schema or annotations, the description covers what is returned, an availability caveat, and the parameter's provenance. Sufficient to call correctly, with only minor gaps around failure/empty-result behavior.

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 0%, so the description must compensate, and it does: it names the sole parameter, ties it to search_legal_literature output, and gives two concrete id format examples ('ncl:A15001353', 'grb:13540821'). This meaningfully exceeds the bare schema.

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 states a specific verb+resource: retrieving the bibliography, abstract, and (when available) full text of research literature. It clearly contrasts with the sibling search_legal_literature by being the id-keyed single-item fetch, so an agent can distinguish it without opening the 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?

It explicitly states the id comes from search_legal_literature, establishing the search-then-retrieve workflow that selects this tool over its search sibling. It gives no explicit when-not or failure conditions, which keeps it from a 5.

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

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

With no annotations, the description carries the full burden and discloses the key behavioral traits: the post-1970 reason-availability constraint and a chained workflow (gazette_pdf_id feeds get_legislative_record for full meeting minutes). It does not cover auth, rate limits, or error behavior, but for a read-only lookup the coverage is solid.

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-loads purpose, then a clear contrast paragraph, then structured Args/Returns blocks. The Returns section is longer than typical but justified because no output schema exists. No wasted filler.

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?

With no output schema, no annotations, and 0% param coverage, the description supplies the missing return shape (versions oldest→newest with date/action/text/reason, source_url, latest_amendment_process) and the parameter meaning. An agent has everything needed to call and interpret it.

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 0%, so the description must compensate, and it does: law_name is explained as a full statute name with abbreviation-to-official-name normalization via 全國法規資料庫, and article_no gets format examples ('184', '15-1'). This adds genuine semantics beyond the bare typed fields.

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: it fetches the text and legislative reasons (立法理由) of a given article across its successive enactments/amendments. It explicitly distinguishes itself from the sibling query_regulation(include_history=True) by explaining that this tool returns the Legislative Yuan's deliberation 'reasons', so an agent can tell them apart without opening schemas.

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?

Names the alternative (query_regulation with include_history=True) and the exact condition that selects this tool instead ('why is this article worded this way / what was the purpose of the amendment'). It also states a hard coverage limit — reasons only exist for amendments after ROC year 59 (1970) — which tells the agent when the tool will not return useful data.

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

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

There are no annotations, so the description carries the behavioral burden. It discloses that the returned content includes the full text plus related items (sponsors, review progress, related documents, public comment deadlines, draft comparison tables), which is meaningful context. However, it says nothing about permissions, access restrictions, or what happens if an unsupported ID prefix is passed.

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?

A single dense sentence plus a short Args block; front-loaded with the verb and resource. It is somewhat long due to the parenthetical breakdown of each record type, but every clause adds useful retrieval context.

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 no annotations, no output schema, and a 1-parameter tool, the description supplies the record types, the id source, and the content the agent can expect to retrieve. It omits only permissions or fallback/error behavior, which would complete the picture.

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 0% for the single parameter, so the description must compensate. It does explain the accepted id formats (bill:…, gazette:…, draft:…, lispdf:…) and their meaning, which is valuable beyond the bare 'record_id' schema string, though it doesn't cover the format for every possible prefix (e.g., what exactly a gazette id looks like).

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 (取得立法紀錄全文) and enables the agent to distinguish it from its siblings by enumerating the supported record_id prefixes (bill, gazette, draft, lispdf) and their contents. It does not mention get_legislative_history except as a producer of its argument, so the boundary with that sibling is only implicit.

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 states where record_id values come from (search_legislative_records, get_legislative_history) and what kinds of records are accepted. It doesn't explicitly state when to use this lookup versus other retrieval tools, but for a lookup-by-id tool the usage context is made clear.

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

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

With no annotations, the description carries the full burden and does well: it discloses the max-50 articles per call, that omitting article_no returns only article_count plus the number range (no text), and that undivided documents return full_text. It omits any pagination guidance for results exceeding 50 and says nothing about permissions or rate limits.

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?

Purpose is front-loaded, followed by return behavior and then arg semantics, all earning their place. Minor redundancy: the 'no article number → range only / full_text' behavior is stated once in the body and again in the Args section.

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?

Without an output schema, the description adequately explains return shapes (articles with number/content, article_count, full_text) and the 50-item cap. The only meaningful gap is what to do when a regulation has more than 50 articles, since no pagination parameter exists.

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 0%, so the description must compensate and does: it explains regulation_id's provenance and gives detailed article_no syntax with single ('15', '15-1', '第十五條之一'), range ('1~10'), and multi ('3,5,15-1') examples, plus the default behavior when blank. This adds substantial meaning beyond the bare string schema.

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 (取得/retrieve) and resource (條文 of 地方自治法規、條約協定、交易所規章), clearly delimiting it from the search-oriented sibling search_other_regulations. An agent can tell this is the fetch-by-id step for non-judgment legal documents.

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?

Explicitly ties regulation_id to the output of search_other_regulations, establishing the retrieval workflow after a search. It does not spell out when-not-to-use or name competing get_* siblings, so routing is implied rather than fully explicit.

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

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

With no annotations, the description discloses important behaviors: covering 11,700+ regulations, supporting fuzzy matching, and returning a dictionary with pcode or suggestions. It lacks details on error handling or rate limits but is adequate for a simple lookup tool.

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 concise with two clear paragraphs, front-loading the purpose and providing Args/Returns sections. 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?

For a simple one-parameter tool with no output schema, the description covers purpose, input, output format, and key behaviors like fuzzy matching and coverage, making it fully self-contained.

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 schema only provides the parameter name and type with 0% coverage. The description adds meaning by explaining law_name as a Chinese regulation name with examples (e.g., 民法, 勞基法), which goes beyond the schema.

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's purpose: converting law names to pcode codes for the national law database. It is distinct from sibling tools like get_citations or get_interpretation, which serve different retrieval functions.

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 implies usage for converting law names to pcode, providing context of coverage and fuzzy matching. However, it does not explicitly state when to use this tool versus alternatives or when not to use it.

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

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

No annotations are provided, so the description carries the full burden. It discloses the return payload (category, fields, full_text, related_laws, attachments, source_url), which is real behavioral information, but it is silent on read-only semantics, permissions/auth, pagination, and error behavior for an invalid id.

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 purpose sentence is front-loaded, and the Args/Returns structure is scannable with no filler. The Returns list is slightly padded with parenthetical examples, but every element is informative.

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?

With no output schema, the Returns section usefully enumerates what comes back, and the single required parameter is fully explained, so an agent has what it needs to invoke and consume the tool. Missing only is any note on failure modes or whether the id must come from the same session's search.

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 0% – 'Precedent Id' is just a bare string – so the description must compensate, and it does: it identifies the parameter's origin (search_precedents output) and gives three concrete id format examples spanning different record categories. It stops short of explaining whether formats are interchangeable, but adds substantial meaning.

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: retrieve the full text (全文) of a precedent. It also explicitly ties itself to search_precedents, so an agent can distinguish it from that sibling (search vs. fetch full text) without opening either 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?

The description establishes the workflow prerequisite by saying the id comes from search_precedents results, which tells the agent when this tool applies. It does not name any exclusions or alternative fetch tools (get_judgment, get_interpretation), but the usage context is clear.

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

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

No annotations exist, so the description carries the full burden, and it does disclose meaningful behavior: a mode-dependent (two-phase) response that depends on whether crime is supplied, the content shape of the statistics, and an interpretive caveat that this is not a sentencing benchmark. Auth, pagination and error behavior are not covered.

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, then coverage, caveat, usage flow, and argument reference in a logical order. It is longer than average but each block (coverage list, workflow, arg formats) carries information the schema does not.

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 6-parameter, no-annotation, no-output-schema tool, the description supplies the return content, the interactive flow, and per-parameter formats. The main remaining gap is that it never explains how it differs from the other statistics siblings in the toolset.

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 0%, so the description must compensate, and it does: crime accepts a name or system code (stole, fraud), law and court accept comma-separated multiples, factors uses the 因子=選項;因子=選項 format with a concrete example, and year_from/year_to are identified as 民國年. All six parameters are given usable syntax.

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 and resource: 司法院事實型量刑資訊系統的刑度統計, and specifies exactly what is returned (符合條件的判決數、各刑種平均／最高／最低與分布). It also names the 10 covered case categories. It does not, however, differentiate itself from sibling statistics tools such as get_statistics or search_statistics.

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?

Explicitly describes the workflow: without crime it lists crime names and courts; with crime it returns law_options, factor_options and current statistics, then narrows with law/court/factors. It also warns the data is past-judgment statistics, not sentencing guidelines. No sibling alternative is named, so it stops short of 5.

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

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

No annotations are provided, so the description carries the full disclosure burden. It does indicate the return content (table contents or the full report text), which is useful, but it says nothing about permissions, size/large-result handling, or behavior on an invalid id. Modest value above structured fields.

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?

One front-loaded purpose sentence followed by a short examples list for the single argument. No filler. Slightly terse but neither padded nor truncated.

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 simple single-parameter retrieval tool with no output schema, the definition is nearly adequate. It does not address result size/pagination, failure modes for malformed ids, or whether the returned table and report-text cases differ, leaving some gaps.

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 0% and the schema only names the property 'Statistics Id' with no description. The description compensates by giving concrete id formats across sources (judicial:, moj:, cprc:, including a nested cprc path), which meaningfully helps an agent construct a valid value.

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 concrete verb+resource: retrieve the content of a statistics table or the full text of a statistics report. It also anchors the input to an id produced by search_statistics, which separates it from the search siblings. It does not, however, explicitly disambiguate from other statistics-family tools such as get_sentencing_statistics.

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 parenthetical '(search_statistics 回傳的 id)' implies the intended workflow: use search_statistics first, then call this with its id. There is no explicit when-to-use/when-not statement or named alternative. Usage is only inferable, not stated.

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

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

With no annotations, the description carries the full burden and does so well: it discloses the 50-article cap, that overflow returns has_more plus a continuation point, that non-existent requested articles appear under missing, that omitting article_no yields structure (headings and article-number ranges) rather than text, and that English translations frequently lag Chinese amendments. Behavioral traits are disclosed far beyond a bare parameter list.

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 core retrieval behavior before the arg list, and each section earns its place. There is mild redundancy between the Args and Returns blocks (article_history and history are each explained twice), which keeps it from the top mark.

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?

Despite no output schema, the description fully specifies the return shape: law (pcode, name, status, plus last_amended, category, effective_date/effective_note), articles, source_url, and the optional history and article_history fields. Together with the documented edge cases, an agent has everything needed to call and interpret this 7-parameter tool.

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 0%, so the description must compensate entirely, and it does: law_name is noted as auto-converted to pcode, pcode's format is exemplified ('B0000001'), article_no's three accepted syntaxes are shown with real values ('184', '184~198', '184,185,247-1'), from_no/to_no are defined as an equivalent range form, and include_history/language carry their own semantics.

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 and resource (retrieve articles from the national regulations database) and enumerates the supported granularities: single article, range, cross-number multiple. However, it never distinguishes itself from the closely-named sibling search_regulations, so an agent must infer which one finds laws versus reads articles.

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-conditions: omit article_no to get only the chapter/article index first, then specify article_no to read text; set include_history=True when the user asks about amendment history; check effective_date/effective_note before citing newly amended articles; use language='en' for English versions. It stops short of naming any alternative tool or when-not-to-use-this-tool case.

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

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

No annotations, so the description carries the full burden and does so richly: real-time querying of official sites, slow responses (監察院 tens of seconds), 律師懲戒 returning 0 when >100 matches, 農業部/教育部 capped at five pages, some sites matching titles only, unmasked personal names in pre-108 / 法務部 / 原民會 cases, and missing dates for 勞動部 results.

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?

Purpose is front-loaded in the first line and the source caveats are organized as scannable lists, but the text is long and dense with source-specific minutiae that an agent rarely needs on every call.

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?

With no output schema and zero schema coverage, the description still documents the return shape (id, agency, category, date, summary plus per-source categories counts and error handling) and the full-text continuation path. Nothing critical for correct invocation is missing.

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 0%, so the description must compensate and it does: keyword (full-text, some sources title-only), source (comma-separated, defaults listed), year_from/year_to (民國年), doc_number (with format examples such as 行政院「A-115-000633」, 工程會「訴1130123」), and page (per-source pagination).

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 (searching administrative appeal decisions and quasi-judicial rulings) and immediately clarifies the distinction from the sibling get_administrative_decision ('要讀全文請把 id 傳給 get_administrative_decision'). An agent can tell exactly which tool returns metadata vs. full text.

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?

Explicitly splits sources into 'default when source is unset' and 'only when source is specified', and gives concrete trigger conditions and caveats per source (e.g. 律師懲戒 needs precise name/case-number; 工程會 has no keyword search, use doc_number). Alternative tool get_administrative_decision is named with the exact handoff parameter.

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

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

With no annotations, the description carries the full burden and delivers: result ordering (newest first), returned fields (id, agency, category, doc_number, date, summary), per-source pagination with differing page sizes, per-category error handling on source failure, and a detailed explanation of the status field semantics including the crucial warning that missing status does not mean still-valid.

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 the sibling alternative, and each section (sources, return format, status semantics, args) earns its place for a tool spanning dozens of sources. The exhaustive agency/source enumeration is heavy, but it is functionally load-bearing because agency behavior depends on it, so the length is defensible rather than padded.

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 highly complex tool with no annotations, no output schema, and 0% schema coverage, the description covers everything an agent needs: source list, default vs named-agency behavior, return shape, pagination quirks, error handling, and status interpretation caveats.

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 0% for all 6 parameters, so the description must compensate and does: keyword (full-text, space-separated, article syntax, IP-office chapter-name limitation), agency (comma-separated with examples and named-only agencies), year_from/year_to (民國年), doc_number (two accepted formats), and page (per-source page sizes). Every parameter is documented with format and behavior.

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 and resource (searching 各機關的行政函釋、解釋令、法規諮詢意見、審查基準) and immediately contrasts itself with the sibling search_interpretations, which covers 大法官解釋/憲判字. An agent can distinguish this from search_interpretations without opening either 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?

Explicitly names the alternative (search_interpretations) and the condition that selects it, plus states that certain agencies (外交部、退輔會、核安會、國發會) are only queried when named, and what happens when agency is omitted. Deduplication behavior across sources is also stated.

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

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

No annotations are provided, so the description carries the full burden, and it does disclose meaningful behavior: results are cached locally for one day, the id can be chained into another tool, and the field set returned differs per status. It does not mention auth requirements, rate limits, or result caps, leaving some gaps.

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 one-line purpose followed by a tight bulleted breakdown of statuses and fields, then args. Efficient, though the separate 'Args' block partly repeats the parameter meanings and adds mild redundancy.

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?

With no annotations, no output schema, and 0% schema description coverage, the description supplies status semantics, per-status return fields, the id linkage, and caching. It still leaves result counts, pagination, and any freshness/auth caveats unstated, so it is good but not fully 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?

Schema coverage is 0%, so the description must compensate, and it largely does: keyword matching scope (case number, petitioner, cause, space-separated terms) and the three allowed status values are documented. It omits the default behavior when status is omitted (schema defaults to 'pending'), which keeps it from a 5.

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 (list) and resource (undecided constitutional court cases) and explicitly contrasts with the sibling get_interpretation, which only holds published rulings. An agent can distinguish this from the other constitutional tools without opening a 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?

Explains the three status buckets (pending / hearing / amicus) and routes the agent onward: the returned id feeds get_constitutional_case_file. It implies the alternative (get_interpretation) but never states an explicit when-not condition or prerequisite, so it stops just 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.

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

No annotations are provided, so the description carries the full behavioral burden. It discloses useful non-obvious behavior: results include case_id for chaining, and include_old only takes effect when year=0 (>0 restricts to 新制憲判字). It does not state result ordering, pagination, or behavior on empty keyword, so it falls short of a 5 for a 7-parameter read tool.

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-loads the purpose and the sibling routing in the first two sentences, then lists args compactly. The Args block is long but each line carries distinct semantics; only mild redundancy from restating defaults already in the schema.

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 parameter-heavy tool with no output schema and no annotations, the description covers scanning scope, filtering semantics, defaults, and the chaining contract. Weakness: it says results carry case_id but not what other fields come back or how results are ordered/limited in practice.

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 0%, so the description must compensate entirely, and it does: every one of the 7 parameters is defined, including the match fields for keyword (標題/字號/爭點/理由書全文), the 民國 year convention, the 0=不篩選 sentinel, inclusive number_from/number_to, the include_old/year interaction, and the max_results default.

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 (列舉大法官解釋/憲法法庭裁判) and the search scope (關鍵字全文搜尋 covering 爭點 + 理由書). It also names the sibling it is not (行政機關的函釋、解釋令 → search_agency_interpretations), so an agent can separate it from the other interpretation tools without opening schemas.

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?

Explicitly gives the when-not case ('行政機關的函釋、解釋令不在這裡，用 search_agency_interpretations') and the downstream contract ('每筆結果帶 case_id，可直接傳給 get_interpretation()'), routing the agent to both the correct alternative and the natural next call.

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

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

No annotations, so the description carries the full burden and does so: result sorting by court authority, the fields returned per hit, main_text win/loss phrase semantics, and an explicit data-coverage caveat (complete only from 2000, sparse 1992-1999, none before 1991) with the instruction to tell users the source does not cover early cases. This is exactly the behavioral context that prevents agent errors.

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?

Long but well organized with bracketed sections, front-loaded purpose, and concrete examples. The keyword-as-citation pattern is explained in two places, but they cover distinct cases (disambiguation vs citation lookup), so the density is largely earned.

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 9-parameter search tool with no annotations and no output schema, the definition covers every parameter, the return dict shape (success, query, total_count, results, cached, timestamp), and the source-coverage limits. Nothing essential for correct invocation is missing.

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 0%, so the description must compensate and it does for all 9 parameters: each is described with examples, underlying field mapping (jud_kw, jud_jmain), required-when conditions for case_word/case_number, and defaults/caps (max_results default 10, limit 200).

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 ('搜尋司法院裁判書系統') and goes further by distinguishing case-number lookup (case_word+case_number) from thematic full-text search (keyword). It does not name sibling tools like get_judgment or search_precedents, so an agent gets the within-tool distinction but no explicit sibling routing.

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?

Explicit when-to-use guidance for two modes, with a concrete worked example ('114年度上易字第503號' → case_word/case_number/year_from). It also gives the when-NOT: putting a case number in keyword turns into full-text search that matches other judgments citing it, and reserves that pattern for citation lookup.

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

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

With no annotations, the description carries the full behavioral burden and does disclose meaningful traits: it names each source's coverage and flags that full text is only sometimes available (作者授權者有全文, 報告全文需在官網下載, 臺大只取可定位卷期). It does not describe result format or pagination behavior, so it stops short of fully transparent.

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 purpose and scope, then a structured source list, then usage/routing and arg notes. It is on the long side, but every block (sources, routing, citation guidance) earns its place rather than padding.

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 multi-source search tool with no annotations and no output schema, the definition covers scope, source semantics, the follow-up retrieval path, and citation formatting. Only return-shape and pagination behavior are left implicit, which is a minor gap given the id→get_legal_literature routing.

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 0%, and the description compensates thoroughly: it defines keyword fields (題名、作者、主題) and the space-separation convention, source semantics with comma handling and empty=all, and critically that years are ROC years (民國年). This adds real meaning the bare schema cannot convey.

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 and resource (搜尋法學研究文獻) and immediately bounds the scope: only official and open-access sources, explicitly excluding paid databases like 月旦/華藝/法源. It also distinguishes itself from the sibling get_legal_literature by naming it as the follow-up for abstracts/full text.

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 (official and open-access literature sources, enumerated one by one) and when-not (paid databases excluded). It also routes the agent to get_legal_literature for the detail step, so the workflow is unambiguous.

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

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

With no annotations, the description carries the full burden and does well: it discloses pagination (20 per page, 10 for drafts), the default term scoping rule ('屆期不連續', term=0 semantics), that gazette is full-text with hit snippets in 'matches', and that each bill carries proposer, date, session, progress and a comparison-table PDF. It does not discuss result ordering, totals, or failure modes.

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 purpose sentence, then a well-organized bulleted breakdown by kind that lets an agent scan to the relevant branch. Slightly long and repeats the kind names in both the list and the Args block, but every section carries information.

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?

With no output schema and no annotations, the description fills the gap by explaining what each result set contains, how pagination works, and how to chain the returned id into get_legislative_record. Minor omissions remain (ordering, total counts, empty-result behavior) but nothing blocking correct invocation.

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 0%, yet the description documents every one of the 5 parameters with value semantics: keyword types, the four kind values, status values split per kind (pending/all/passed vs pending/closed), term meaning including the special 0 behavior, and page. This fully compensates for the empty schema descriptions.

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 ('搜尋立法動態與立法紀錄') and then enumerates the four distinct record families (bills, gazette, drafts, join) with their content, which makes the tool's scope unambiguous and clearly distinct from siblings like search_regulations and get_legislative_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?

Gives concrete selection guidance: which kind to pick for which need, a specific retrieval recipe for gazette ('查立法者原意時可用「法律名稱＋條次」'), and routes to get_legislative_record for full text via the returned id. It lacks explicit 'do not use this for X' exclusions against the many sibling search tools, so not a full 5.

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

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

With no annotations, the description carries the full behavioral burden and does well: local regulations are current-only, some treaties are scanned images with PDF links only, exchange regulations come from 證基會 and are view-only/non-redistributable, and the default fans out to all 28 sources. Return-format and pagination behavior are not described, keeping it from 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?

Front-loaded with the purpose and the sibling it does not replace, then organized into scannable bullets by source category with the args block last. It is longer than minimal but each section carries routing or coverage information, so little is wasted.

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 no annotations and no output schema, the description supplies the source taxonomy, the default fan-out, key caveats, and the id-handoff to get_other_regulation. An agent has enough to invoke it correctly; only pagination/result-shape details are absent.

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 0%, so the description must compensate, and it does for two of three params: keyword gets syntax guidance and an example, and source gets a full list of accepted values plus delimiter rules and default behavior. page is only glossed as 頁數 with no detail, leaving a small gap against the otherwise strong coverage.

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 (搜尋) and resource, and explicitly scopes itself against a named sibling: it finds regulations that query_regulation cannot. The three bulleted categories (地方自治法規, 條約及協定, 交易所規章) make the coverage boundary concrete and let an agent distinguish it from search_regulations/query_regulation without opening any 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?

Explicitly routes the agent: it names query_regulation as the source it complements and tells the caller to pass the returned id to get_other_regulation. It also advises specifying source and warns that omitting it queries all 28 sources. It stops short of stating when-not to use it, so a 4 rather than 5.

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

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

With no annotations, the description carries the full burden and does so well: it discloses the per-category page size (20), the site's 500-record cap per category, pagination behavior, and the meaning of result fields such as reference_value, status, status_note, and 編註. It omits auth/access prerequisites but otherwise provides unusually rich operational 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 definition is long but well-structured: purpose is front-loaded, then a bulleted taxonomy, then caveats, then limits. Given the category list effectively substitutes for a missing enum in the schema, the length is largely earned, though a few lines could be tightened.

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 complex, domain-specific taxonomy tool with no annotations and no output schema, the description supplies the category definitions, default behavior, result caps, pagination guidance, and field semantics an agent needs. The absence of an output schema is well compensated by the explanation of reference_value/status/編註 fields.

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 0%, so the description must compensate, and it does: keyword is defined as full-text search, category accepts comma-separated values with a documented default set, and year_from/year_to are clarified as 民國年. Only page is left to inference, a minor 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?

States a specific verb (搜尋) and resource (司法院判解資料), then explicitly carves out its scope by noting these are the categories search_judgments cannot return. The enumerated category list lets an agent distinguish it from siblings without opening any 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?

Explicitly routes the agent away from search_judgments and lists which categories belong here, plus the default category set when none is supplied. It also gives mitigation advice (add keyword or year when results are too numerous) and points to get_precedent for 編註, though it does not state general when-not-to-use conditions.

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

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

With no annotations, the description carries the full burden and does well: it discloses the corpus size (11,700+ 部), page size (50 per page), the exact return fields, and the ordering rule that switches based on amended_since. It stops short of stating read-only nature, rate limits, or permission needs, but the behavioral profile is otherwise strong.

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 purpose sentence followed by clearly labeled Args and Returns blocks; each parameter line earns its place with a concrete example. Slightly dense with parenthetical examples, but nothing is redundant filler.

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 five-parameter search tool with no output schema and no annotations, the description supplies everything needed: scope, page size, ordering semantics, per-parameter meaning, and the return fields. Nothing an agent needs to invoke it correctly is missing.

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 0%, so the description must compensate, and it does for all five parameters: keyword with omission rule and examples, offset as pagination start, exclude_abolished with its default and the caveat that abolished laws remain searchable but flagged, amended_since with dual date-format examples (2026-09-01 / 115-09-01), and category with the matching column and examples.

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 and resource (搜尋法規名稱 / 列出某日之後新制定或修正的法規) and clearly frames two operating modes including compliance tracking. It does not distinguish itself from close siblings like query_regulation, search_other_regulations, or get_other_regulation, so an agent cannot route between them from this text alone.

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 in-tool conditions (keyword may be omitted when amended_since or category is supplied; amended_since triggers newest-first ordering) and names the compliance-tracking use case. However it never says when NOT to use it or which sibling to prefer for adjacent needs, so alternatives remain inferred.

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

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

With no annotations, the description carries the full burden and does disclose meaningful behavior: the keyword matches titles only, source options are enumerated, the year is a ROC year, and the two-step retrieval flow plus the pipe-delimited output format of get_statistics are explained. It omits pagination behavior, result ordering/limits, and any auth/permission 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?

Purpose is front-loaded on the first line, followed by a compact source list, the result-handling flow, and a clean per-argument breakdown. Slightly verbose in the source enumeration, but every element is informative and non-redundant given the empty schema.

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 4-parameter tool with no annotations and no output schema, the description covers parameter meaning, source scope, and the retrieval handoff to get_statistics. It is largely sufficient, with only minor gaps around paging semantics and result-size expectations.

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 0%, so the description must compensate, and it documents all four parameters: keyword (matches title), source (with candidate values and 'blank = all'), year (ROC, 'blank = latest'), and page. Only 'page' is left vague (page count/size unstated), so it nearly but not fully compensates.

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 (搜尋) and resource (官方法律統計表與報告) and enumerates the exact corpora it covers (司法統計年報/月報、法務統計、犯罪狀況及其分析). An agent can distinguish it from search_judgments or search_regulations without opening any 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?

The description establishes the workflow explicitly: result ids are passed to get_statistics to retrieve the table text or full report, which clarifies when to use this tool versus the retrieval sibling. It does not state exclusions or when another statistics tool (e.g. get_sentencing_statistics) is preferable, so it stops short of full routing guidance.

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