Claude for Legal

Anthropic's legal plugin suite: contracts, privacy, employment, IP, litigation, AI governance and legal education

PluginEditors’ pickMedium risk

anthropics/claude-for-legal

Install

/plugin marketplace add anthropics/claude-for-legal
/plugin install commercial-legal@claude-for-legal
/commercial-legal:cold-start-interview

Replace commercial-legal with your practice: privacy-legal, corporate-legal, employment-legal, ip-legal, litigation-legal and more. Restart Claude Code before the interview.

This is third-party code. Review the repository files before installing.

What it does

The repository holds practice-area plugins: commercial contracts, privacy, product counsel, corporate deals, employment, regulatory, AI governance, IP, litigation, legal clinics and law students. Each plugin starts with a cold-start interview that learns your playbook and seed documents and writes a practice profile every skill reads. Named agents then handle specific jobs: NDA and vendor agreement review, DSAR responses, termination review, trademark clearance, case chronologies, deposition prep and deadline tracking. Output is framed as drafts for attorney review, with source attribution and explicit gates before anything is sent.

Who it is for. For in-house and firm lawyers, legal clinics and law students working with US and EU law.

Good fit when

  • You need to review a contract against your playbook and brief the business
  • You need to track renewal, leave or IP registration deadlines
  • You need to classify an AI system under the EU AI Act or draft a privacy impact assessment

Not a fit when

  • The work is under a jurisdiction the skills are not built for
  • You need a final legal position without attorney review

Example request

Review this NDA against our playbook and flag clauses for escalation

Limitations

Skills are built around US and EU law and US legal practice. All output is a draft, not legal advice. The cold-start interview takes 10-20 minutes per plugin. Without a connected legal research tool citations are unverified, and many connectors need paid accounts.

How to disable. Remove the plugin via /plugin; the practice profile lives in ~/.claude/plugins/config/claude-for-legal/.

Security check

  • Handles confidential legal documents
  • Connectors reach external systems such as DocuSign and Slack

README in short

The README stresses that all output is a draft for attorney review and does not reflect Anthropic's legal positions. It covers practice-area plugins, managed-agent cookbooks for scheduled work, MCP connectors to Slack, Google Drive, DocuSign, CourtListener and others, plus a table of named agents with commands. Install via Cowork or the Claude Code marketplace, then run the cold-start interview and connect a legal research tool.

FAQ

Why is the output generic?

Most likely the cold-start interview was skipped; without a practice profile the skills do not know your playbook.

Can I run it outside Claude Code?

Yes, install the plugins in Claude Cowork or deploy them via the Claude Managed Agents API using the cookbooks.

Editors’ pick

Microsoft's utility and MCP server that turn PDF, Word, Excel, PowerPoint, HTML and other files into Markdown for models

MCP serverMedium risk184K
Editors’ pick

Anthropic's official skill examples: docx, pdf, pptx and xlsx documents, design, MCP builder and the Agent Skills spec

SkillMedium risk176.3K
Editors’ pick

A skill that builds editable PPTX decks from PDF, DOCX, URLs and Markdown with native shapes, charts and animations

SkillMedium risk54.4K

PaddleOCR agent skills

PaddleOCR Agent Skills

Two official PaddleOCR skills: text recognition from images and PDFs, and parsing complex documents into Markdown

SkillMedium risk89.5K