Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add lawchat-oss/taiwan-legal-plugin --skill interpretation-lookupgit clone --depth 1 https://github.com/lawchat-oss/taiwan-legal-pluginWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/lawchat-oss/taiwan-legal-plugin/interpretation-lookup)<a href="https://agentmods.dev/skills/lawchat-oss/taiwan-legal-plugin/interpretation-lookup"><img src="https://agentmods.dev/badge/skills/lawchat-oss/taiwan-legal-plugin/interpretation-lookup/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/lawchat-oss/taiwan-legal-plugin/interpretation-lookup"><img src="https://agentmods.dev/badge/skills/lawchat-oss/taiwan-legal-plugin/interpretation-lookup.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00214 | $0.02876 |
| Opus 5.5 | $0.00086 | $0.01150 |
| Sonnet 5.5 | $0.00043 | $0.00575 |
| Haiku 4.5 | $0.00021 | $0.00288 |
Grade A, and why
interpretation-lookup scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 17d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 109 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/taiwan-legal:interpretation-lookup
Finds and retrieves interpretive authority: constitutional interpretations, agency interpretations, and court resolutions / Q&A conferences / precedents.
Audience
Legal researchers, attorneys, paralegals, in-house counsel, and law students working in or with the Taiwanese legal system. Assumes familiarity with the weight of each source (釋字 binding on all organs; 函釋 binding on the agency but reviewable by courts; 決議 and 判例 lost their special binding force in the 2019 Grand Chamber reform). Not designed for end-consumers seeking legal advice.
Instructions
-
Load practice profile. Read
~/.claude/plugins/config/taiwan-legal-plugin/taiwan-legal/CLAUDE.mdfor the user's citation style. If the file does not exist, say: "Run/taiwan-legal:cold-start-interviewfirst to set defaults — or shall I proceed with no defaults?" Do not apply the profile's judgment-search date window here: 決議 predate the 2019 reform and 判例 / 院字・院解字 are decades old, so a "last 5 years" window would exclude them. -
Scope check. Confirm the request is a lookup or research query, not an advice request (see "What this skill does NOT do" below).
-
Identify intent and pick the tool.
- 釋字 / 憲判字 by number →
get_interpretation(case_id=...); addreasoning_keyword=orinclude_reasoning=truefor the reasoning,include_opinions=truefor Justices' opinions. - Constitutional topic →
search_interpretations(keyword=...); what a ruling cited →get_citations(case_id=...); which later 釋字 / 憲判字 cited it →get_citations(case_id=..., direction="cited_by"). - Case-file materials (聲請書, 答辯書, 鑑定意見, 法庭之友意見書, 言詞辯論筆錄, 爭點題綱) →
get_constitutional_case_file(case_id=...), optionally withkeyword=to find which filings discuss a point; read one withdocument_id=. Cases not yet decided →search_constitutional_docket(keyword=..., status="pending" | "hearing" | "amicus"). - How an agency reads a provision / 函釋 on a topic →
search_agency_interpretations(keyword=..., agency=...). Leaveagencyempty to search every source; use the agency's name (e.g. 勞動部, 財政部, 金管會, 銓敘部, 人事總處, 消保處, 地政司, 臺北市, 新北市) when the user names one; 外交部, 退輔會, 核安會, 國發會, NCC, 客委會, 僑委會, 運動部, 關務署新頒釋函 and 陸委會主站廣告函釋 are searched only when named. A known 字號 →doc_number=.... Thenget_agency_interpretation(interpretation_id=...)for the full text. Patent / trademark examination guidelines:agency="智慧局"with a chapter term such as 專利要件 or 混淆誤認. - 決議 / 法律問題座談 / 判例 / 院字・院解字 / 大法庭 →
search_precedents(keyword=..., category=...), thenget_precedent(precedent_id=...). - Curated judgments with 裁判要旨 →
search_precedents(keyword=..., category="精選裁判"); only those a court designated 具參考價值 / 足資討論 →category="具參考價值裁判"(items carryreference_value). - Administrative appeal and quasi-judicial decisions →
search_administrative_decisions(keyword=..., source=...), thenget_administrative_decision(decision_id=...). Withoutsourceit searches 行政院訴願, 公平會, 不當勞動行為裁決, 保訓會, and 金管會裁罰. Name the source for 工程會採購申訴審議判斷 (source="採購申訴"; no keyword search upstream — use the case number such as 訴1130123 or a year), 監察院 (slow), 律師懲戒 (needs a precise keyword), 醫事懲戒 (currently posted medical disciplinary notices; matches names, locations and certificate numbers), or a ministry's / local government's 訴願 (e.g.source="臺北市", orsource="訴願"for all of them). 內政部, 衛福部 and 勞動部 訴願 and 醫事懲戒 are off by default: they come back withdisabled: trueand asource_url— give the user that link instead of retrying. Decisions come back as the official sites publish them; older Executive Yuan (filed up to 民國 108) and 法務部 decisions and 原民會 decisions show the parties' names. Executive Yuan cases filed up to 民國 108 use the 院臺訴字 number as their id (e.g.ey:1070210137). - Years are ROC years (民國; 2026 = 115).
- 釋字 / 憲判字 by number →
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 17d ago Changed 805ced7512be
- 1mo ago Changed · +1 lines · +3 tokens per session aca1f23804b9
- 1mo ago First seen · 108 lines · 211 tokens per session scan A 77ca9e2b494b
interpretation-lookup is a skill published in the GitHub repository lawchat-oss/taiwan-legal-plugin (43 stars, last pushed 18d ago), licensed MIT. It adds 214 tokens to every session and 2,876 once invoked, about $0.0009 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-03.
interpretation-lookup compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| interpretation-lookuplawchat-oss/taiwan-legal-plugin | Skill | 43 | 18d ago | MIT | A | 214 | Claude Code |
| fto-triageanthropics/claude-for-legal | Skill | 9.9k | yesterday | Apache-2.0 | A | 81 | Claude Code |
| legal-holdanthropics/claude-for-legal | Skill | 9.9k | yesterday | Apache-2.0 | A | 68 | Claude Code |
| specification-writingwanshuiyin/Auto-claude-code-research-in-sleep | Skill | 19k | today | MIT | A | 49 | Claude Code |
| regulatory-research-fallbackHKUDS/OpenSpace | Skill | 7.9k | 2d ago | MIT | A | 20 | Claude Code, Codex |
| terms-of-servicegenspark-ai/genoffice | Skill | 12k | today | Apache-2.0 | A | 32 | Claude Code, Codex |
| x-scorecardx-cmd/x-cmd | Skill | 4.7k | 8d ago | Apache-2.0 | A | 57 | Claude Code, Codex |