Claude for Legal: Complete Setup Guide for Lawyers (2026)

What Is Claude for Legal?

Claude for Legal is Anthropic's official open-source legal toolkit, launched May 12, 2026. It bundles 12 practice-area plugins, more than 80 workflow agents, and 20+ connectors to major legal research and document platforms, all running through Claude's desktop agent (Cowork) or Claude Code. It gives attorneys citation-aware drafting, jurisdiction-specific research, and document analysis tuned for legal work. If you bill by the hour, that changes your advantage on every matter.

It is not a separate product and carries no extra charge. It is free to use on any paid Claude plan, layered on top of your existing subscription. Think of it as switching Claude from general assistant to legal associate.

Why Claude for Legal Matters for Law Firms

General-purpose AI produces general-purpose output. Claude for Legal fixes three problems attorneys hit with base Claude:

• Citation handling: The toolkit's workflow agents format citations in Bluebook, ALWD, or jurisdiction-specific style and flag citations that need verification, reducing the hallucination risk that has led to sanctions against unprepared attorneys.
• Document analysis at scale: Upload contracts, depositions, or discovery sets running hundreds of pages. The agents index the documents and answer questions with pinpoint references — page, paragraph, and clause number.
• Jurisdiction awareness: Set your state and practice area once. Every response accounts for local rules, statutes, and procedural requirements.

The 2025 Clio Legal Trends Report found that 79% of lawyers who adopted legal-specific AI tools reported higher client satisfaction scores than those using general AI tools. Claude for Legal exists because legal work demands precision that generic prompts cannot deliver.

Step-by-Step Setup: Account to First Prompt in 10 Minutes

Step 1: Choose Your Plan

Claude for Legal is free to use on any paid Claude plan. Solo attorneys can run it on Pro ($20/month), and firms handling client data should run it on Team (standard seats $20/user/month billed annually) or Enterprise (custom pricing) for the commercial data terms. For a full plan comparison, see our Claude AI pricing guide.

Choose Team if you are a solo practitioner or small firm handling client data. Choose Enterprise if your firm requires SSO, SCIM provisioning, or SOC 2 Type II compliance documentation. The 2025 Thomson Reuters Generative AI Report found that only 41% of legal organizations have established AI use policies — firms handling sensitive client data should choose Enterprise tiers that include SOC 2 compliance, SSO, and audit logging.

Step 2: Create Your Workspace

Go to claude.ai/team and create a workspace. Name it after your firm. Add attorneys as members. Each member gets their own conversation history — no cross-contamination of client data between users.

Step 3: Install Claude for Legal

Open Claude's desktop agent (Cowork) or Claude Code and install the Claude for Legal toolkit. It is open source, so you can review exactly what each component does. During setup you will configure three things:

• Practice-area plugins: Pick from the 12 practice-area plugins (corporate, litigation, family law, real estate, IP, etc.). This tunes Claude's output for your work.
• Workflow agents: Enable the workflow agents you need from the 80+ available, such as citation formatting (Bluebook, ALWD, or state-specific), contract review, and deposition summaries. Set your primary state and federal circuit so Claude applies them unless you specify otherwise in a prompt.
• Connectors: Connect the major legal research and document platforms your firm already uses, so Claude can pull from and file to them directly.

Step 4: Configure Security and Confidentiality

This step is non-negotiable. Under Workspace Settings → Security:

• Run client work on a Team or Enterprise plan. Under Anthropic's commercial terms, your data is not used for model training by default and Anthropic acts as a data processor, with retention controls and agreements available. This is the baseline for ABA-compliant AI use.
• Enable audit logging if your firm tracks who uses AI and when.
• On Enterprise, configure SSO/SAML so attorneys sign in through your firm's identity provider.

ABA Formal Opinion 512 states: "A lawyer must make reasonable efforts to prevent the inadvertent or unauthorized disclosure of, or unauthorized access to, information relating to the representation." Commercial plans with no-training-by-default terms satisfy this requirement. On consumer plans, you risk an ethics violation every time you paste client data into Claude.

Your First Three Legal Prompts

Once Claude for Legal is active, try these three prompts to see what it can do. Each uses the CRAFT framework for optimal results.

Prompt 1: Contract Red-Flag Scan

Upload a contract and use this prompt:

"Review this agreement as a [practice area] attorney in [state]. Identify all clauses that create risk for [party]. Flag any non-standard terms, missing protections, or provisions that deviate from market norms. Format your findings as a table with columns: Clause, Section, Risk Level (High/Medium/Low), and Recommended Action."

The contract review agent returns a structured table with section references. In a peer-reviewed study, experienced attorneys averaged 92 minutes to review five NDAs while AI completed the same task in 26 seconds with higher accuracy, according to LawGeex's NDA review benchmark. Structured prompts like this one close the gap even further.

Prompt 2: Research Memo Draft

"Draft a research memorandum on [legal question] under [state] law. Include: (1) a brief answer, (2) the governing statutory framework, (3) key case law with full citations, (4) analysis of how the law applies to [specific facts], and (5) a conclusion with recommended next steps. Use Bluebook citation format."

The citation agent formats every reference and flags any citation it cannot verify with high confidence. You still verify — but the starting point is far better than a general AI draft.

Prompt 3: Deposition Summary

"Summarize this deposition transcript. For each topic discussed, provide: the topic heading, page-line references, key admissions or contradictions, and any objections made. Highlight testimony that supports or undermines [specific claim]."

For a deeper dive on deposition workflows, see our deposition summary guide.

How Claude for Legal Compares to Harvey AI and CoCounsel

Three AI tools dominate legal tech in 2026. Here is how they differ:

Feature	Claude for Legal	Harvey AI	CoCounsel (Thomson Reuters)
Base AI model	Claude (Anthropic)	Multi-model (includes Claude)	Multi-model
Context window	Up to 1M tokens on current models	Varies by model	Varies by model
Training on your data	No, by default (Team & Enterprise)	No (enterprise terms)	No (enterprise terms)
Pricing	Free on paid Claude plans (Team seats from $20/user/mo)	Custom (reported ~$1,200/user/mo entry)	Bundled with Westlaw
Citation verification	Flags low-confidence cites	Links to Westlaw via partnership	Integrated with Westlaw
Self-serve setup	Yes	No (sales process)	No (enterprise sales)
Solo/small firm access	Yes	Limited	Westlaw subscription required

The 8am 2026 Legal Industry Report found that 69% of legal professionals now use generative AI — more than double the prior year — with individual practitioners adopting far faster than firms. Claude's self-serve model and lower price point make it a natural fit for solo and small firm attorneys. Harvey and CoCounsel remain strong choices for Am Law 100 firms with existing Westlaw or enterprise AI budgets.

Security Best Practices for Law Firms

Setup is only half the job. Maintain these practices to stay compliant:

• Written AI policy: Every firm needs one. Define which plans are approved, what data can be entered, and how output must be verified.
• Client disclosure: At least 14 state bars now require or recommend telling clients when AI is used in their matter. Check your jurisdiction's guidance.
• Prompt hygiene: Never paste opposing counsel's privileged communications into Claude. Redact sensitive identifiers when possible.
• Output verification: Treat every Claude response as a first draft. Verify citations, check calculations, and confirm factual claims. The plugin helps — but final responsibility stays with you.