本文へ移動
cccskills
無料GitHub で公開

originality-in-european-copyright-joris-deene

a skill that determines whether a given subject matter qualifies for copyright protection under EU law by applying the Cofemel two-step test (concept of work + originality), grounded in fifteen CJEU judgments (including the most recent Mio/Konektra and Calinescu) and the four EU directive provisions on originality. It argues from one of two positions: pro-rightsholder (establishing that the work is original) or pro-alleged-infringer (contesting work status or originality via the four exclusion grounds — technical function, rule-dictated outcome, sweat of the brow, idea/functionality). It includes sector-specific modules for photography, software/GUI, sporting events, databases, applied art, functional texts, derivative works/critical editions, and multimedia works.

インストール方法を見る

含まれるファイル(29)

  • SKILL.md11.4 KB
  • case-law/brompton-c-833-18.md23.8 KB
  • case-law/bsa-c-393-09.md14.4 KB
  • case-law/calinescu-c-649-23.md32.1 KB
  • case-law/cofemel-c-683-17.md21.6 KB
  • case-law/fapl-c-403-08.md13.3 KB
  • case-law/football-dataco-c-604-10.md16.6 KB
  • case-law/funke-medien-c-469-17.md15.4 KB
  • case-law/infopaq-c-5-08.md11.0 KB
  • case-law/levola-c-310-17.md18.6 KB
  • case-law/mio-konektra-c-580-23-c-795-23.md36.5 KB
  • case-law/nintendo-c-355-12.md14.8 KB
  • case-law/painer-c-145-10.md12.9 KB
  • case-law/renckhoff-c-161-17.md8.0 KB
  • case-law/sas-institute-c-406-10.md21.3 KB
  • case-law/sony-datel-c-159-23.md18.9 KB
  • methodology/00-work-concept.md12.8 KB
  • methodology/01-originality-test.md52.2 KB
  • methodology/02-photography.md10.6 KB
  • methodology/03-software.md28.9 KB
  • methodology/04-databases.md12.1 KB
  • methodology/05-applied-art.md26.0 KB
  • methodology/06-text-and-fragments.md10.0 KB
  • methodology/07-derivative-works.md19.5 KB
  • methodology/08-multimedia-works.md12.1 KB
  • positions/pro-alleged-infringer.md91.5 KB
  • positions/pro-rightsholder.md63.3 KB
  • README.md9.1 KB
  • workflow.md4.3 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Skill: Originality and Concept of Work in European Copyright (CJEU)

Purpose

To build a methodical analysis of the question whether a concrete object (text, photograph, software, database, utilitarian article, sound recording, design, …) is protected by copyright under EU law, based on the case law of the Court of Justice of the European Union. The skill produces a developed argumentation from one of two positions:

  • Pro-rightsholder — argue why the object qualifies as a work and is original
  • Pro-alleged-infringer — argue why the object does not qualify as a work or is not original

Version 1.0 — first public release. Integrates the EU statutory harmonization framework for originality: verbatim citations of the four directive provisions that explicitly address originality:

  • Art. 1(3) Directive 2009/24/EC (computer programs — "author's own intellectual creation" + "no other criteria")
  • Art. 3(1) Directive 96/9/EC (databases — "author's own intellectual creation" by "selection or arrangement" + "no other criteria")
  • Art. 6 Directive 2006/116/EC (photographs — "author's own intellectual creation" + "no other criteria" + optional protection of "other photographs")
  • Art. 14 Directive (EU) 2019/790 (DSM) (works of visual art in the public domain — anti-reviving regime)

These provisions constitute the statutory floor of the Cofemel two-step test. The common "no other criteria shall be applied" clause in three of the four provisions is the doctrinal basis for the Cofemel rule (C-683/17, paragraph 29) that Member States may not impose additional requirements. The skill makes the doctrinal connection between primary EU law and CJEU case law explicit in a dedicated section "The EU harmonization framework for originality" in 01-originality-test.md. Verbatim texts are also included in the relevant sectoral modules (02-photography.md, 03-software.md, 04-databases.md, 07-derivative-works.md).

The doctrinal core consists of fifteen key CJEU judgments and one footnote judgment (Top System): Infopaq (C-5/08), Painer (C-145/10), BSA (C-393/09), FAPL (C-403/08 & C-429/08), Football Dataco (C-604/10), SAS Institute (C-406/10), Levola Hengelo (C-310/17), Cofemel (C-683/17), Brompton (C-833/18), Funke Medien (C-469/17), Renckhoff (C-161/17), Sony/Datel (C-159/23), Mio/Konektra (C-580/23 + C-795/23), Călinescu/FNSA (C-649/23) and Nintendo/PC Box (C-355/12), with Top System (C-13/20) as a footnote mention.

Scope

Within scopeOutside scope
Whether an object is protected by copyrightInfringement analysis (reproduction, communication to the public)
The originality requirement under EU lawTerm of protection
The EU concept of work (objective identifiability)Exceptions and limitations (quotation, parody, …)
Sector-specific refinements based on CJEU case lawNeighbouring rights
Two positions: pro-rightsholder and pro-alleged-infringerCopyright contracts and licensing
Sui generis database right
Design protection as such (but: cumulation with copyright via Cofemel test is in scope)
National implementation in Member States (this skill addresses EU-level harmonization only; national law must be consulted separately)

The two-step test

The CJEU has progressively built the EU concept of work. Since Cofemel (C-683/17, paragraph 29) the concept of work is explicitly tested in two cumulative steps.

Step 1 — Objective identifiability (concept of work in the narrow sense)

Is the object sufficiently precisely and objectively identifiable? It must be an expression that is recognisable and unambiguously delimitable for third parties, even though this expression need not be permanent (Levola Hengelo, C-310/17, paragraph 40). Subjective, sensory or non-objectifiable experiences (taste, smell, fleeting impression) do not qualify (Levola operative part). The objectivity requirement serves a dual legal-certainty function: both the authorities responsible for enforcement and the third parties against whom protection is invoked must be able to clearly and precisely identify which object is protected (Cofemel paragraphs 32-33).

Full treatment in methodology/00-work-concept.md.

Step 2 — Originality (author's own intellectual creation)

Is the object original in the sense that it is the author's own intellectual creation (Infopaq, C-5/08, paragraph 37; consolidated in Cofemel paragraph 29)? It must be the expression of free and creative choices that reflect the author's personality (Painer, C-145/10, paragraphs 88-92), not dictated by technical function (BSA, C-393/09, paragraphs 48-49; Brompton, C-833/18), external rules (FAPL, C-403/08 + C-429/08), convention or banality.

Operational sub-tests:

  • "Own" — originating from the author, not derived
  • "Intellectual creation" — the fruit of mental effort; no "sweat of the brow" (Football Dataco, C-604/10; Funke Medien, C-469/17)
  • "Selection, arrangement and combination" in an original manner (Infopaq paragraph 45)
  • Free and creative choices that reflect personality through a personal touch (Painer)
  • No exclusion ground: not dictated by technical function, rules, sweat of the brow or idea/functionality

Full treatment in methodology/01-originality-test.md.

Both steps are cumulative: an object must satisfy both to qualify as a copyright-protected work (Cofemel paragraph 29).

File structure

originality-in-european-copyright/
├── SKILL.md                                       (this file)
├── methodology/
│   ├── 00-work-concept.md                        [Levola Hengelo + confirmatory cross-references to Cofemel, Brompton, Funke Medien, Mio/Konektra, Călinescu, Nintendo]
│   ├── 01-originality-test.md                    [all 15 judgments + EU harmonization framework (art. 1(3) Dir. 2009/24, art. 3(1) Dir. 96/9, art. 6 Dir. 2006/116, art. 14 Dir. 2019/790 DSM) + Cofemel consolidation + Brompton + Funke Medien + Sony/Datel + Mio/Konektra uniqueness requirement + Călinescu anti-severability + skill irrelevant + public-domain safeguard]
│   ├── 02-photography.md                         [Painer + Renckhoff + Cofemel + verbatim art. 6 Dir. 2006/116]
│   ├── 03-software.md                            [BSA + SAS Institute + Sony/Datel + Nintendo outer limit + Top System footnote + verbatim art. 1(3) Dir. 2009/24]
│   ├── 04-databases.md                           [Football Dataco + verbatim art. 3(1) Dir. 96/9]
│   ├── 05-applied-art.md                         [Cofemel + Brompton + Mio/Konektra (three consolidated key judgments)]
│   ├── 06-text-and-fragments.md                  [Infopaq + Funke Medien + brief mention of Călinescu with cross-reference to 07]
│   ├── 07-derivative-works.md                    [Călinescu/FNSA as key judgment + verbatim art. 14 Dir. 2019/790 DSM anti-reviving regime; room for future case law on translations/adaptations/arrangements]
│   └── 08-multimedia-works.md                    [Nintendo/PC Box as key judgment for complex matter; room for future NFT/AR-VR/AI-multimedia case law]
├── case-law/
│   ├── infopaq-c-5-08.md
│   ├── painer-c-145-10.md
│   ├── bsa-c-393-09.md
│   ├── fapl-c-403-08.md
│   ├── football-dataco-c-604-10.md
│   ├── sas-institute-c-406-10.md
│   ├── levola-c-310-17.md
│   ├── cofemel-c-683-17.md
│   ├── brompton-c-833-18.md
│   ├── renckhoff-c-161-17.md
│   ├── funke-medien-c-469-17.md
│   ├── sony-datel-c-159-23.md
│   ├── mio-konektra-c-580-23-c-795-23.md
│   ├── calinescu-c-649-23.md
│   └── nintendo-c-355-12.md
├── positions/
│   ├── pro-rightsholder.md                       [concept of work + 8 sectors + horizontal lines of argument on the EU statutory floor (four argument lines A-D)]
│   └── pro-alleged-infringer.md                  [concept of work + 8 sectors + horizontal lines of attack on the EU statutory floor (four lines of attack F-I)]
└── workflow.md

Methodology

  1. Identify the object and the sector. What is it concretely (text, photograph, software, utilitarian article, …)? Which sectoral case law is relevant?
  2. Identify the position. Is the user arguing from the standpoint of the author/rightsholder, or from the standpoint of the alleged infringer? In case of doubt: ask explicitly.
  3. Consult the relevant case-law files in case-law/ for the applicable judgments.
  4. Apply the two-step test (concept of work → originality) according to methodology/.
  5. Develop the argumentation following the corresponding position file in positions/.
  6. Anticipate the counter-argument from the other position.

For the detailed procedure see workflow.md.

Anti-hallucination rule

This skill cites exclusively from judgments for which a reference file exists in case-law/. For each judgment referenced in the analysis, the corresponding file must exist and contain the cited paragraph. For other CJEU judgments or for national case law: write "case reference to be added" and provide only the legal principle.

Citation form

Standard form for CJEU judgments:

CJEU [date], C-[case number], [short name], ECLI:[ECLI].

For example: CJEU 16 July 2009, C-5/08, Infopaq International, ECLI:EU:C:2009:465.

Output form

Standard inline in chat. For developed legal argumentation (procedural document, advice, formal notice): offer a markdown artifact.

Limitations

  • Version 1.0 — covers EU-level harmonization only. National implementation in each Member State is outside scope and must be consulted separately by the practitioner.
  • No factual assessment of the work itself — the skill provides the legal argumentation structure; the factual assessment of whether an object is objectively identifiable or displays original choices remains the responsibility of the user and ultimately of the court.
  • Strict scope — the skill addresses only the concept of work (step 1) and originality (step 2). Exploitation rights (reproduction, communication to the public, distribution, modification right — including the question left unanswered in Sony/Datel), exceptions and limitations (art. 5 Directive 2001/29, decompilation under art. 6 Software Directive), anti-circumvention of TPMs (art. 7(1)(c) Software Directive, art. 6 InfoSoc Directive) and the interplay with fundamental rights fall outside scope and belong in separate skills.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Creates an Agent Authority Charter for enterprise or regulated AI agents before deployment. Use this Skill when a user needs to define what an AI agent is allowed to do, who delegated authority to it, what actions are permitted or prohibited, when human approval is required, what evidence must be preserved, and how the agent can be suspended, revoked, or escalated.

日本語の概要は準備中です。原文の説明を表示しています。

lawve-ai/awesome-legal-skills8632026年10月3日 更新

EU AI Act Knowledge Engine — authoritative regulatory Q&A grounded in 70 official EU source documents (including the 2026 Commission draft guidelines on Art. 6 high-risk classification — general principles, Annex I, Annex III). Answers any question about the AI Act with article-level citations from the full regulation text, Commission guidelines, EDPB/EDPS opinions, codes of practice, harmonised standards, FRIA guides, incident templates, and sector-specific guidance. This skill should be used when the user asks about the "EU AI Act", "AI Act regulation", "Regulation 2024/1689", "KI-Verordnung", asks to "explain an AI Act article", "what does Article X say", "AI Act requirements", "AI Act penalties", "AI Act timeline", "GPAI obligations", "high-risk AI", "prohibited AI practices", "AI Act and GDPR", "fundamental rights impact assessment", "AI Act standards", "AI Act national implementation", "AI literacy", "serious incident reporting", "Digital Omnibus", "AI regulatory sandbox", or any question about EU artificial intelligence regulation. Also triggers on German terms: "KI-Gesetz", "KI-Verordnung", "Hochrisiko-KI", "KI-Kompetenz", "Grundrechte-Folgenabschaetzung", "GPAI-Verhaltenskodex". For assessment workflows that produce a classification decision rather than a knowledge answer, offer to run a structured risk-tier classification in-session.

日本語の概要は準備中です。原文の説明を表示しています。

lawve-ai/awesome-legal-skills8632026年10月3日 更新

This skill builds a structured audit trail of an AI-assisted task: what the tool was asked to do, what materials it was given, what it produced, how the output was verified, and what was ultimately relied upon. It documents the workflow for supervision and later review without ruling on privilege or disclosure, so the record stays accurate rather than self-serving.

日本語の概要は準備中です。原文の説明を表示しています。

lawve-ai/awesome-legal-skills8632026年10月3日 更新

Use this skill when the user wants an AI governance, legal-risk, privacy, compliance, procurement, or vendor-risk review of an internal AI use case, an AI product feature, an LLM workflow, or a third-party AI vendor. The skill asks intake and clarifying questions first when facts or evidence are missing, identifies required documentation and missing evidence, maps the use case to AI governance frameworks and applicable legal domains, and produces a preliminary or final governance review with scorecards, findings, owners, remediation actions, and follow-up questions.

日本語の概要は準備中です。原文の説明を表示しています。

lawve-ai/awesome-legal-skills8632026年10月3日 更新

Turn an interpretive-ambiguity audit of a legal text — contract, statute, regulation, or judicial opinion — into a polished deliverable. Produces a multi-page website (default), a single-page interactive site, a Microsoft Word document, a PowerPoint deck, or a LaTeX/PDF. Use whenever a user has a stress-test result, ambiguity audit, or "where will this be litigated" scenarios and wants to publish, present, or share them. Triggers on "publish the audit," "make a report from this," "turn this into a website / deck / brief / Word doc," "make a Netlify site," "build slides of these ambiguities," "render the scenarios," "package this stress test," "produce a LaTeX version," and similar. Pairs with the ambiguity-stress-test skill (the underlying analysis) but works on any structured or semi-structured ambiguity analysis. Visual design is opinionated but the user can override colors, fonts, and brand label. Most formats need Python and a writable filesystem; single-file HTML works where the host has neither.

日本語の概要は準備中です。原文の説明を表示しています。

lawve-ai/awesome-legal-skills8632026年10月3日 更新

Adversarially stress-tests a legal text — a contract, statute, regulation, or judicial opinion — for interpretive ambiguity: it finds the seams where the people governed by it will later disagree about what it means and turns each into a concrete dispute scenario with both sides' arguments, the likely outcome, and a fix. Use it whenever someone wants to pressure-test, red-team, audit, or find weak spots, loopholes, gaps, ambiguities, or drafting problems in a legal document; whenever a drafter wants to tighten a contract, statute, regulation, or opinion before it issues; whenever a litigator wants to mine an opinion or contract for arguments; or whenever someone hands over a legal text and asks where it will be fought over or for issue-spotting. Trigger it for contract review, statutory-ambiguity analysis, judicial-opinion scope analysis, and drafting QA — even if the user never says "stress-test" or "ambiguity."

日本語の概要は準備中です。原文の説明を表示しています。

lawve-ai/awesome-legal-skills8632026年10月3日 更新

lawve-ai のスキルをすべて見る

このスキルの問題を報告する