Klotzkette/claude-fuer-deutsches-recht

⚠️ Experimentelle Skill-Sammlung für deutsches Recht (Arbeits-, Gesellschafts-, Insolvenz-, Datenschutz-, Prozessrecht u.a.) – inzwischen verbessert und im Alltag getestet, aber weiterhin Experiment. Bitte selber ausprobieren, Issues/PRs willkommen! Keine Rechtsberatung. Mandatsgeheimnis (§§ 203/204 StGB, § 43e BRAO), DSGVO, US-Transfer, KI-VO & Co

📚 Quick Overview

What it is – A massive, community‑maintained collection of Claude‑compatible plugins, skills and prompt templates for German‑law practice. The repo bundles ready‑to‑install plugin ZIPs (each covering a legal area), fine‑grained skills (single‑step workflows), and self‑contained Markdown prompts ("Werkstatt" = detailed, "Mini" = compact) that can be used without any installation.

Who it helps – Law firms, in‑house legal teams and consultants who want to harness Claude (or any LLM that supports the ChatGPT‑plugin marketplace) for tasks such as:

  • reviewing contracts, notices or court filings,
  • drafting legal documents (contracts, pleadings, memos, etc.),
  • performing structured legal analysis, and
  • training junior staff with realistic test‑cases.

How you use it – Choose a legal domain, install the corresponding plugin via the ChatGPT desktop or workspace UI, or simply download a Markdown prompt and feed it to Claude with your case files. Test‑cases ("Testakten") are provided for safe experimentation.

Why it matters – German law is highly formalised and subject to strict data‑privacy rules. This collection gives practitioners a ready‑made, jurisdiction‑specific prompt library that respects the new AI‑Regulation (KI‑MIG) and German data‑protection standards, while clearly warning that the output is not legal advice.

🔧 Core Components (as described in the README)

Component What it is When to use it Where to find it
Plugin ZIP Installable package containing a whole legal area (skills, references, helper files). When you want a persistent, reusable setup for a specific domain. Listed in the Asset Index (ASSET_INDEX.md).
Skill A single, narrowly scoped workflow (e.g., “check contract clause X”). For a one‑off step inside a larger process. SKILLS.md and per‑plugin pages under skills-index/.
Werkstatt‑Prompt Detailed, multi‑step Markdown prompt that can run alone. Complex cases where you need depth but do not want to install a plugin. docs/werkstatt-und-schnellstart-coverage.md → Werkstatt‑Prompts.
Mini‑Prompt Compact prompt for quick start, limited to ~7 500 characters. Fast, lightweight tasks or prototyping. Same doc as above → Schnellstart‑Prompts.
Testakte Synthetic case files (PDF/ZIP) for safe trial runs. Practicing the workflow without exposing real client data. testakten/README.md.

🚀 Getting Started (step‑by‑step)

  1. Pick a domain – Browse the Plugin‑Katalog (section “Was ist drin”) to see a short description of each legal area.
  2. Choose an entry point –
    • Plugin: follow INSTALLATION_EINFACH.md and import the marketplace JSON into ChatGPT (desktop → Settings → Import or Workspace → Admin → Plugins → Add).
    • Prompt: download a Werkstatt or Mini prompt from the coverage doc and paste it into Claude with your documents attached.
  3. Provide the case material – upload the contract, notice, or other file the prompt/skill asks for.
  4. Run the workflow – Claude will ask follow‑up questions, produce a draft, and you can iterate until the final document is ready.
  5. Validate – The README stresses a mandatory legal‑review before any real use.

📂 Repository Size & Structure

  • ≈300 plugins (each a ZIP).
  • ≈23 278 skills – fine‑grained markdown workflows.
  • ≈298 Werkstatt‑Prompts and ≈297 Mini‑Prompts.
  • 517 central test cases (plus additional plugin‑specific ones).
  • Navigation aids: auto‑generated alphabetic tables, asset index, changelog, and a dedicated English quick guide.

⚠️ Important Disclaimers (from the README)

  • Not legal advice – the outputs are experimental and must be reviewed by a qualified lawyer.
  • Potential errors – statutes may be outdated; the repo does not guarantee completeness.
  • Data‑privacy – using the plugins may involve file uploads; ensure compliance with GDPR and the German AI‑Regulation.
  • Liability – the authors waive liability where legally permissible.

📖 Where to Learn More

  • Installation guide – INSTALLATION_EINFACH.md
  • Quick start – QUICKSTART.md
  • Full skill list – SKILLS.md
  • English quick guide – section #english-quick-guide
  • Changelog & releases – linked at the top of the README.

All information above is taken directly from the repository’s README; no additional features have been inferred.

Related

  • Project
    borghei/Claude-Skills

    Claude Skills is an open‑source library of 385 ready‑to‑use “skills” – prompt‑plus‑Python bundles – that can be loaded into Claude, ChatGPT, Gemini, Copilot, Cursor and other LLM assistants. It ships 77 expert personas, 900 pure‑Python tools, and compliance workflows for 18 regulated frameworks, letting non‑engineers run real business, product‑management and compliance tasks via slash‑commands or plugin installation.

  • Project
    feiskyer/claude-code-settings

    A set of Claude Code plugins (skills & sub‑agents) that add brainstorming, deep research, GitHub issue/PR automation, image generation, translation, and more, with ready‑made config files for many LLM providers.

  • Project
    jamditis/claude-skills-journalism

    A library of Claude Code (and limited Codex) plugins/skills that give AI agents journalism‑specific capabilities—AP‑style enforcement, fact‑checking, FOIA templates, PDF design, video analysis, research tools, etc.—installable via Claude’s marketplace or by copying `SKILL.md` folders.

  • Project
    coreyhaines31/makerskills

    makerskills is an open‑source Claude Code plugin that bundles 22 AI‑driven “skills” – ready‑to‑run workflows for decisions, research, knowledge‑base management, content creation, finance, and automation. Users invoke a skill with a slash command (e.g., `/decide`), the plugin runs the defined LLM prompts and any needed external tools, and stores structured results in a personal config folder. Designed for solo founders and small teams, it offers composable, documented workflows that can be extended via `skillify`, `toolify`, and `loopify`.

  • Project
    luwill/research-skills

    A set of five Claude Code plug‑ins that automate academic research tasks—exhaustive literature search, citation‑integrity‑driven review writing, proposal drafting, and two kinds of slide‑deck generation (stylised images vs. fully editable, vector‑based decks). All tools enforce strict citation checks and are installable as simple skill folders.