Tw Legal RAG
aa0101181514/tw-legal-rag
Retrieve real Taiwan court judgments with verifiable citations before answering any question about Taiwan law or case law.
First-pass framework for reading, interpreting, and structuring statutory analysis of US federal, state, and local law.
$ npx skills add lawve-ai/awesome-legal-skills --skill statute-guide -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install lawve-ai/awesome-legal-skills statute-guide --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/lawve-ai/awesome-legal-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/statute-analyzer-rafal-fryc .claude/skills/statute-guide && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "statute-guide" agent skill from https://github.com/lawve-ai/awesome-legal-skills/tree/main/skills/statute-analyzer-rafal-fryc into .claude/skills/statute-guide/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "statute-guide", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/lawve-ai/awesome-legal-skills/tree/main/skills/statute-analyzer-rafal-frycType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add lawve-ai/awesome-legal-skills --skill statute-guide -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install lawve-ai/awesome-legal-skills statute-guide --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/lawve-ai/awesome-legal-skills.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/statute-analyzer-rafal-fryc .agents/skills/statute-guide && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "statute-guide" agent skill from https://github.com/lawve-ai/awesome-legal-skills/tree/main/skills/statute-analyzer-rafal-fryc into .agents/skills/statute-guide/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "statute-guide", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add lawve-ai/awesome-legal-skills --skill statute-guide -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install lawve-ai/awesome-legal-skills statute-guide --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/lawve-ai/awesome-legal-skills.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/statute-analyzer-rafal-fryc .cursor/skills/statute-guide && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "statute-guide" agent skill from https://github.com/lawve-ai/awesome-legal-skills/tree/main/skills/statute-analyzer-rafal-fryc into .cursor/skills/statute-guide/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "statute-guide", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/lawve-ai/awesome-legal-skills.git --path skills/statute-analyzer-rafal-fryc--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add lawve-ai/awesome-legal-skills --skill statute-guide -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install lawve-ai/awesome-legal-skills statute-guide --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/lawve-ai/awesome-legal-skills.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/statute-analyzer-rafal-fryc .gemini/skills/statute-guide && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "statute-guide" agent skill from https://github.com/lawve-ai/awesome-legal-skills/tree/main/skills/statute-analyzer-rafal-fryc into .gemini/skills/statute-guide/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "statute-guide", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install lawve-ai/awesome-legal-skills statute-guideInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add lawve-ai/awesome-legal-skills --skill statute-guide -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/lawve-ai/awesome-legal-skills.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/statute-analyzer-rafal-fryc .github/skills/statute-guide && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "statute-guide" agent skill from https://github.com/lawve-ai/awesome-legal-skills/tree/main/skills/statute-analyzer-rafal-fryc into .github/skills/statute-guide/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "statute-guide", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add lawve-ai/awesome-legal-skills --skill statute-guide -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install lawve-ai/awesome-legal-skills statute-guide --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/lawve-ai/awesome-legal-skills.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/statute-analyzer-rafal-fryc .opencode/skills/statute-guide && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "statute-guide" agent skill from https://github.com/lawve-ai/awesome-legal-skills/tree/main/skills/statute-analyzer-rafal-fryc into .opencode/skills/statute-guide/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "statute-guide", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
statute-guideFirst-pass framework for reading, interpreting, and structuring statutory analysis of US federal, state, and local law.
Statute Guide is an agent skill from lawve-ai/awesome-legal-skills. First-pass framework for reading, interpreting, and structuring statutory analysis of US federal, state, and local law. Produces draft analysis for attorney review — not legal advice. Use this skill whenever the user references a specific US statute, regulation, ordinance, or rule by citation, asks "what does [statute X] require," asks for compliance scoping, applicability thresholds, requirement extraction, exemption analysis, definitional analysis, federal preemption analysis, or multi-state comparison — even…
Its SKILL.md is about 6.5k tokens, which your agent loads only when the skill is triggered. The skill folder holds 8 other files, including reference files (for example `CHANGELOG.md`, `README.md` and `references/canons_of_construction.md`).
It sits in Legal & Compliance, covering Legal research and Citation management. The repository describes itself as: A curated list of awesome Agent Skills for automating legal work. The licence is AGPL-3.0.
3 steps, taken from the step headings in SKILL.md.
Read from SKILL.md and the folder at commit 045f738. It shows what the files ask for, not the result of running them.
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
No scripts in the folder and no shell commands in SKILL.md.
From the folder's file list and the shell code blocks in SKILL.md.
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Statute Guide loads about 6.5k tokens when it runs, and up to ~14k if it reads all its reference files. Until then it costs about 168 tokens; SKILL.md has 2,807 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
The full file from lawve-ai/awesome-legal-skills at commit 045f738, republished under its AGPL-3.0 licence (© lawve-ai). 2,807 words, ~6,523 tokens.
.claude/skills/statute-guide/SKILL.md (or your agent's skills folder). This skill also uses 7 other files; get the full folder from GitHub.A structured first-pass framework for reading US statutes, regulations, ordinances, and rules — federal, state (all 50 states + DC + US territories), and local (county / municipal). Claude organises the analysis; a named attorney owns every legal conclusion.
This skill does not handle non-US law (EU, UK, Canada, foreign jurisdictions, international treaties). If the user's matter touches foreign law, halt and route to counsel qualified in that jurisdiction.
Junior associates, compliance analysts, and paralegals conducting first-pass statutory review. All outputs require attorney review before any client-facing, regulatory, or transactional use. This skill is not for self-represented users seeking legal advice — the firewall between "structuring analysis" and "giving advice" exists because misreading statutory text has real consequences for real people, and a named attorney is the only person qualified to bear that risk.
Accretive Judgment. Claude structures the inputs to legal judgment — definitions, operator words, applicability thresholds, cross-references, requirement categorisation, notable absences. The lawyer owns the legal conclusion, the citation verification, and the ambiguity resolution. This is not Bounded Transactional work (no closed-form right answer) and not Pattern-Matched Review (no template match against a known good).
Use when the user asks for first-pass structuring of:
This skill does not perform:
If a request falls into any of the above, halt and route to a qualified attorney rather than attempting a first-pass — these areas have their own evidentiary, doctrinal, and process requirements that a statutory-text framework cannot satisfy. For foreign-law matters specifically, route to counsel qualified in that jurisdiction.
Before starting Step 1, confirm all of the following are present. Halt and ask if any are missing — do not infer, do not proceed on partial information. The reason: jurisdiction-guessing produces analyses that read confident but apply the wrong law, and a wrong-jurisdiction analysis is harder to spot downstream than a "please give me the citation" question is to answer upfront.
| Required Input | What it must contain |
|---|---|
| Statutory citation | Jurisdiction + section number (e.g., "Cal. Civ. Code § 1798.100"; "42 U.S.C. § 1320d-2"; "Chicago Municipal Code § 4-4-005") |
| Full text or authoritative link | Consolidated/codified version, not a summary |
| Jurisdiction level | Federal, state (which state — all 50 + DC + territories), or local (which county / city / municipality, plus the parent state) — explicit, not inferred |
| Version / effective date | Which version of the statute is being analysed |
| Stated purpose | Compliance scoping, element extraction, exemption check, or other |
Halt-and-ask rule: If the user provides only a citation without text, only text without citation, or an ambiguous jurisdiction (e.g., "the privacy law" without naming the state), stop and request the missing element. Do not infer jurisdiction from context clues; do not proceed on partial citations.
Non-US flag: If the citation, text, or context indicates a non-US source (e.g., "Regulation (EU) 2016/679", "Data Protection Act 2018", "PIPEDA", a foreign court reporter), halt and decline — this skill covers US law only. Recommend counsel qualified in that jurisdiction.
Claude produces a draft analysis, never a final conclusion. Every output must:
DRAFT — Attorney Review Required | Reviewing Attorney: <name> | Privilege Status: <status><reviewing attorney: ___> placeholder the user must fill)Conclusory verbs without a Confidence Band create the appearance of a final legal conclusion the lawyer never signed off on — tag every conclusion with a band, or downgrade it to an open question for the attorney. Phrases like "Mandatory obligation," "Enforcement by Agency X," or "This applies to your client" stated flatly are exactly the failure mode this section exists to prevent.
Every material finding in the output is tagged High / Medium / Low:
| Band | When |
|---|---|
| High | Text unambiguous; definitions self-contained; no contested agency interpretation; no relevant cross-references unresolved |
| Medium | Cross-references resolved but require verification; minor definitional gaps; agency guidance consistent with text |
| Low | Undefined terms; contested provision; agency interpretation diverges from text; split case law in implementing regs; effective date uncertain |
Flat-confidence outputs are not acceptable for ambiguous provisions. When in doubt between bands, downgrade — the downside of an under-confident finding (lawyer spends extra minutes verifying) is much smaller than the downside of an over-confident one (lawyer trusts a finding that was actually shaky).
Halt the analysis and surface to the named attorney when any of the following occur. Escalation is mandatory; it overrides the instinct to complete the analysis, because finishing-anyway is how shaky readings become "the analysis said so."
When escalating, state which trigger fired and what is needed from the attorney.
| Mode | Why it happens | Guard |
|---|---|---|
| Advice vs. support drift | A confident, structured draft reads like advice; readers stop noticing the disclaimer | Draft label + Delegation Threshold section in every output |
| Privilege leakage | Compliance analysis gets pasted into emails or shared documents without privilege framing, exposing client communications to discovery | If analysis touches a client matter, mark "Attorney Work Product — Privileged & Confidential" and confirm distribution scope before sharing |
| Accountability gap | Output is consumed as a final conclusion because no one is named as owner | Named-attorney requirement; conclusory verbs paired with Confidence Bands |
| Cite-staleness | Statutes amend; an analysis dated six months ago may apply repealed text | Version-date in Input Requirements; Confidence Band caveat naming the version analysed |
| Jurisdiction mismatch (US-internal) | Framework patterns from one state get misapplied to another; federal-law assumptions get carried into state analysis; home-rule assumptions made in a Dillon's Rule state | Jurisdiction level (federal / state / local) confirmed in Input Requirements; one declared jurisdiction per analysis; preemption flagged for attorney |
| Out-of-scope foreign law | User pastes EU / UK / foreign text without saying so; framework's US-specific canons (preemption, agency deference, rule of lenity) get applied to law where they don't exist | Non-US flag in Input Requirements; halt and decline rather than attempt |
| Type | Created By | Characteristics |
|---|---|---|
| Statute | Legislature | Formal enactment; commands, prohibits, or declares policy; provides framework |
| Regulation | Government agency | Implements statutes; has force of law; usually more operationally detailed |
| Rule | Agency or court | Administrative rule (=regulation) or procedural requirement |
Key insight: Statutes give agencies authority to create regulations. Always read both the statute and implementing regulations — regulations often contain the operational details the statute left to agency discretion.
The statute as passed is often not the statute as implemented — a statute with a future effective date may be amended before taking effect.
The same statutory language can mean different things depending on who enforces it.
Browse the index or table of contents before diving into a specific section. Understanding how the issue fits in the larger whole prevents misreading a provision as freestanding when it depends on a definition or applicability section elsewhere.
For unfamiliar statutes or codes, see references/statutory_structure.md — read it when first browsing a statute you have not seen before.
Every word has meaning. Find the definitions section first and reference it constantly.
Build a reference sheet of key definitions before analysing substantive provisions.
Statutes are dense. Every word and punctuation mark has meaning. Read each sentence multiple times; parse complex sentences into their component parts; do not skim. Statutory language rewards close attention and punishes skimming with confident misreads.
These words have consistent legal functions across statutes:
| Term | Meaning |
|---|---|
| Shall | Mandatory — REQUIRED |
| May | Permissive — ALLOWED |
| And | Conjunctive — ALL elements must be satisfied |
| Or | Disjunctive — ANY ONE element is sufficient |
| Unless / Except | Signals an exception to the general rule |
| Subject to | Limited by another section |
| Notwithstanding | Applies DESPITE other sections |
| If...then / Upon / Provided that | A precondition must be satisfied |
| Means | Exhaustive definition follows |
| Includes | Examples follow (may not be exhaustive) |
Misreading "and" as "or" or "shall" as "may" fundamentally changes a provision's meaning — these are the top source of Low-confidence findings and the cheapest mistakes to avoid by reading slowly.
When you encounter a reference to another statute or section, stop and read it. Cross-references may expand, limit, or modify the provision being analysed. Build a map of how sections relate.
When language is ambiguous, use the established interpretive tools below. Ambiguity itself is an escalation signal — flag for the attorney even while applying the tools.
Textual canons (read the text on its own terms):
Purpose canons (read the text in light of its goal):
For full canon definitions with case examples, read references/canons_of_construction.md when interpreting ambiguous statutory language or resolving definitional gaps.
Preamble/purpose clauses, findings sections, and structural context often state legislative intent more cleanly than the operative provisions do.
| Type | Examples | Implementation Team |
|---|---|---|
| Disclosure | Privacy notices, warning labels, terms | Legal/policy |
| Operational | Response deadlines, internal processes | Compliance |
| Technical | System requirements, security standards | Engineering |
| UI/Design | Link placement, font size, button design | Product/design |
Separating these matters: a "privacy policy requirements" checklist should not include operational deadlines that never appear in the policy itself. Separate WHAT must be disclosed from HOW the business must operate.
Determine WHO must comply before extracting requirements.
| Threshold Type | Examples |
|---|---|
| Revenue | Annual gross revenue > $X million |
| Volume | Process data of > X consumers/transactions |
| Revenue from Activity | Derive X% of revenue from regulated activity |
| Entity Type | Applies to developers / controllers / operators |
Conjunctive vs. disjunctive: "$25M revenue AND 100K consumers" is far more limited than "$25M revenue OR 100K consumers." This single word change can multiply or divide the regulated population by an order of magnitude.
The skill handles three US jurisdiction levels and the interactions between them. Each analysis stays within one declared jurisdiction at a time — multi-state or federal/state comparison is a separate pass, done one jurisdiction at a time, then assembled.
| Level | Examples | Typical citation form |
|---|---|---|
| Federal | U.S. Code, Code of Federal Regulations | 42 U.S.C. § 1320d-2, 45 C.F.R. § 164.502 |
| State | State codes, state administrative codes (all 50 + DC + territories) | Cal. Civ. Code § 1798.100, Va. Code Ann. § 59.1-578, N.Y. Gen. Bus. Law § 899-bb |
| Local | County and municipal codes, ordinances | N.Y.C. Admin. Code § 20-870, Chicago Municipal Code § 4-4-005 |
Federal law preempts conflicting state and local law in some areas (express preemption, field preemption, conflict preemption). State law preempts conflicting local law where the state legislature has occupied a field or expressly limited home-rule authority. Preemption analysis is always an escalation trigger — flag for the attorney; do not declare preemption status as a finding.
When comparing similar laws across states (e.g., state privacy laws, breach-notification statutes, consumer protection acts):
In home-rule states, municipalities may regulate areas the state has not preempted. In Dillon's Rule states, municipal authority is narrower. If a local ordinance conflicts with state law, the local provision may be void — flag for attorney rather than declaring.
For detailed lessons from multi-statute analysis (drawn from US state privacy-law work), see references/practical_lessons.md — read when scoping multi-state or federal-vs-state work, or when applicability thresholds look unusual.
| Factor | Questions |
|---|---|
| Authority | AG only? Private parties? Agency? |
| Penalties | Civil / criminal / administrative? Amount? |
| Cure period | Opportunity to fix before penalties? |
| Private right of action | Can individuals sue? |
| Enforcement history | Is the agency actively enforcing? |
Two requirements with identical language can have vastly different practical priority depending on enforcement dynamics — note this in the output, do not treat all statutory text as equally load-bearing.
Compare against typical provisions to identify notable absences. The absence of a remedy or protection is often as significant as what is included.
If an interpretation creates an internal contradiction or an absurd result, it is probably wrong — statutes are written to be coherent. When in doubt, prefer the reading that furthers the statute's stated purpose. (When the absurdity is on the statute's side, not the interpretation's, that is an escalation trigger.)
Every analysis produced by this skill follows this structure. Conclusory verbs without a Confidence Band are not permitted.
DRAFT — Attorney Review Required
Reviewing Attorney: <name>
Privilege Status: <e.g., "Attorney Work Product — Privileged & Confidential" OR "Not privileged">
Statute: <citation> | Version/Effective Date: <date>
Jurisdiction Level: Federal | State (<which>) | Local (<which, parent state>)
Purpose of Analysis: <stated purpose>
1. Applicability
- Threshold(s) identified: <list> [Confidence: High/Medium/Low]
- Conjunctive/Disjunctive: <which>
- Exemptions potentially in play: <list> [Confidence: ...]
- Federal preemption status: <flagged for attorney — do not state here>
- Inter-state / federal-state conflict noted: <yes/no — if yes, list for attorney>
2. Definitions Used
- <Term>: <statutory definition> [exhaustive / illustrative]
3. Operative Provisions (Draft Reading)
- <Provision>: <plain reading> [Confidence: ...]
Operator words: shall/may/and/or as parsed
Cross-references resolved: <list>
Cross-references unresolved: <list> ← escalation candidate
4. Requirements by Type
- Disclosure: <items>
- Operational: <items>
- Technical: <items>
- UI/Design: <items>
5. Enforcement Posture (Descriptive, Not Predictive)
- Authority: <agency / private right>
- Penalty structure: <civil/criminal/admin, amount>
- Cure period: <yes/no, length> [Confidence: ...]
6. Notable Absences
- <items>
7. Open Questions for Attorney
- <Every Low-confidence finding listed here>
- <Every escalation trigger that fired, with which trigger>
- <Citation verification items>
- <Currency-of-text verification>
This document is a draft analytical framework, not legal advice.
No legal conclusion is final until the named attorney signs off.Input prompt (user):
"Can you walk me through Cal. Civ. Code § 1798.105 — the CCPA deletion right? We're scoping a CCPA compliance project. Here's the consolidated text: [paste]. Effective date 2023 amendments. Purpose: element extraction for our policy team."
Output (excerpt — illustrative, not authoritative):
DRAFT — Attorney Review Required
Reviewing Attorney: <placeholder — to be filled by user>
Privilege Status: <to confirm — likely Attorney Work Product if client matter>
Statute: Cal. Civ. Code § 1798.105 | Version: 2023 amendments
Jurisdiction Level: State (California)
Purpose of Analysis: Element extraction for policy team
1. Applicability
- Right runs to "consumers" (defined § 1798.140); duty runs to "businesses"
meeting § 1798.140(d) thresholds [Confidence: High]
- Conjunctive/Disjunctive: applicability uses OR thresholds in § 1798.140(d) [Confidence: High]
3. Operative Provisions (Draft Reading)
- § 1798.105(a): consumer "shall have the right to request that a business delete
any personal information about the consumer which the business has collected
from the consumer." [Confidence: High]
Operator words: "shall have the right" (mandatory right); "from the consumer"
(limits scope to first-party data — see § 1798.105(d) exceptions)
- § 1798.105(d): nine enumerated exceptions to the deletion obligation.
[Confidence: Medium — "necessary to" language in (d)(1)–(d)(9) is operationally
ambiguous and has divergent agency guidance]
7. Open Questions for Attorney
- Verify 2023 amendment version against current CPPA-published text
- Confirm scope of "collected from the consumer" vs. inferred/derived data —
contested area, regulation interpretation diverges from some practitioner views
(Escalation trigger #4: term defined in part by case law / agency practice)The example shows the structural pattern: Confidence Bands on every finding, the first-party-data scope limitation flagged as Medium because the regulations have created ambiguity around it, and a concrete escalation trigger written into the Open Questions block.
Before producing output, confirm:
The references/ directory holds depth content. See references/index.md for the full guide. Quick map:
canons_of_construction.md — Read when interpreting ambiguous textpractical_lessons.md — Read when scoping multi-statute or multi-jurisdiction workstatutory_structure.md — Read when first browsing an unfamiliar statute or codeSee CHANGELOG.md next to this file for version history.
© lawve-ai, AGPL-3.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
SKILL.md and 7 other files (references) in skills/statute-analyzer-rafal-fryc of lawve-ai/awesome-legal-skills.
Open the folder on GitHubat commit 045f738
Statute Guide next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Statute Guide this skilllawve-ai/awesome-legal-skills | 842 | — | ~6.5k | Automated safety check: Pass | AGPL-3.0 | |
| Tw Legal RAGaa0101181514/tw-legal-rag | 328 | — | ~580 | Automated safety check: Pass | Custom licence | |
| Case RetrievalTHUYRan/Legal-Skills-Chinese | 873 | — | ~18k | Automated safety check: Pass | None | |
| Design Award SearchSeanJ1ang/design-judge-skills | 712 | — | ~3k | Automated safety check: Pass | Apache-2.0 | |
| China Lawyer AnalystCSlawyer1985/china-lawyer-analyst | 196 | — | ~3.3k | Automated safety check: Pass | None | |
| Billing And Litigation BudgetTHUYRan/Legal-Skills-Chinese | 873 | — | ~5k | Automated safety check: Pass | None |
aa0101181514/tw-legal-rag
Retrieve real Taiwan court judgments with verifiable citations before answering any question about Taiwan law or case law.
THUYRan/Legal-Skills-Chinese
Trigger this skill when the user needs to find similar cases, related judgments, or adjudicative rules relevant to a current legal issue.
SeanJ1ang/design-judge-skills
Find and verify award-winning designs in the same or adjacent functional category through eight explicit relevance dimensions: problem and user, core function, sensing technology, intervention…
CSlawyer1985/china-lawyer-analyst
通过中国法律视角分析事件,运用成文法解释、指导案例参照、请求权基础分析等方法, 理解权利义务、评估责任风险、识别法律依据并推荐合规策略。
THUYRan/Legal-Skills-Chinese
A skill your agent uses when the user needs to track or manage attorney hours, expert fees, and investigation costs; control litigation spend; or prepare timesheets or expense statements for clients.
Golden2002/legal-research-skill
中国法律全领域系统性检索专家. An agent skill from Golden2002/legal-research-skill.
lawve-ai/awesome-legal-skills
U.S. An agent skill from lawve-ai/awesome-legal-skills.
lawve-ai/awesome-legal-skills
Practitioner skill for advising on EU Regulation 2023/2854 (Data Act).
lawve-ai/awesome-legal-skills
Calendar litigation and arbitration deadlines from a scheduling order.
lawve-ai/awesome-legal-skills
Read, search, and download emails and attachments from Microsoft Outlook via OAuth2.
lawve-ai/awesome-legal-skills
Turn an interpretive-ambiguity audit of a legal text — contract, statute, regulation, or judicial opinion — into a polished deliverable.
lawve-ai/awesome-legal-skills
Audits a website for compliance with Azerbaijan's Law on Personal Data No.
Categories
First-pass framework for reading, interpreting, and structuring statutory analysis of US federal, state, and local law. Statute Guide is an agent skill from lawve-ai/awesome-legal-skills. First-pass framework for reading, interpreting, and structuring statutory analysis of US federal, state, and local law.
Statute Guide fits situations like: the user references a specific US statute; rule by citation; asks what does [statute X] require; asks for compliance scoping.
Run `npx skills add lawve-ai/awesome-legal-skills --skill statute-guide -a claude-code`. Or copy the skill folder (skills/statute-analyzer-rafal-fryc in lawve-ai/awesome-legal-skills) into .claude/skills/statute-guide in your project. Claude Code loads it when a task matches its description.
Run `npx skills add lawve-ai/awesome-legal-skills --skill statute-guide -a codex`. Or copy the skill folder (skills/statute-analyzer-rafal-fryc in lawve-ai/awesome-legal-skills) into .agents/skills/statute-guide in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add lawve-ai/awesome-legal-skills --skill statute-guide -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/statute-guide, .gemini/skills/statute-guide, .github/skills/statute-guide and .opencode/skills/statute-guide in your project.
SKILL.md names no scripts, command-line tools or credentials: Statute Guide is instructions for the agent only.
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
Statute Guide is published under the AGPL-3.0 licence (from the LICENSE file in the skill folder). It allows redistribution, so the full SKILL.md is shown on this page.
About 6.5k tokens (SKILL.md is roughly 26k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 7.1k tokens, read only when the agent opens those files.
Skills that share tags, products or a category with Statute Guide: Tw Legal RAG (aa0101181514/tw-legal-rag, 328 stars), Case Retrieval (THUYRan/Legal-Skills-Chinese, 873 stars), Design Award Search (SeanJ1ang/design-judge-skills, 712 stars) and China Lawyer Analyst (CSlawyer1985/china-lawyer-analyst, 196 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
lawve-ai (a GitHub organization) maintains it in lawve-ai/awesome-legal-skills, which has 842 GitHub stars. The repository holds 154 skills in this directory. The repository was last updated on October 2, 2026.
Source: lawve-ai/awesome-legal-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.