(How to Create a Claude Skill or Plugin for Law and Use It in Claude Cowork)
On February 3, 2026, a single product announcement from Anthropic wiped approximately $285 billion in market capitalization off the stock market in a single trading day. Thomson Reuters dropped 16%. LegalZoom cratered nearly 20%. RELX, the parent company of LexisNexis, fell 14%. Wolters Kluwer lost 13%. The London Stock Exchange Group plunged 8%. The contagion spread to Salesforce, ServiceNow, FactSet, and dozens of other enterprise software companies. Bloomberg called it a “$285 billion rout.” Analysts started calling it the “SaaSpocalypse.”
The catalyst? Anthropic released a set of open-source plugins for its Claude Cowork tool. One of them was a legal plugin that could review contracts, flag risks, triage NDAs, and track compliance. Investors took one look and decided the entire SaaS business model was cooked.
Here’s the part that still gets me: the legal skill at the core of this market panic is, at its heart, a structured markdown file. We’re talking about roughly 250 lines of well-organized pseudo-code, plain text instructions that tell Claude how to think about legal workflows. No compiled binaries. No proprietary algorithms. No infrastructure. Just markdown. And it’s included in the $20/month Claude Pro subscription. That’s less than most attorneys spend on lunch. The full Business Insider breakdown of the stock carnage is here: https://www.businessinsider.com/anthropic-cowork-legal-plugin-publishing-stocks-legalzoom-thomson-reuters-relx-2026-2
So What Are Claude Skills & Plugins, Exactly?
A Claude Skills are a set of instructions, typically written in markdown, that teaches Claude how to approach a specific domain or task in a repeatable, structured way. Think of it as giving Claude a playbook. Instead of prompting from scratch every time you need a contract reviewed or a compliance check run, you write the instructions once, and Claude follows them consistently. Skills include things like domain knowledge (“here’s how our firm handles risk assessment”), step-by-step workflows (“when reviewing an NDA, check these clauses in this order”), and output formatting (“flag issues as green/yellow/red with recommended language”). They’re stored as plain files, organized in folders, and loaded automatically when relevant. Claude Plugins are a bundle that packages one or more skills together with slash commands, MCP connectors, and sub-agents into a single installable unit, while a skill is just a standalone markdown file that teaches Claude how to handle a specific task or domain.
The important thing to understand is that you don’t have to write these from scratch. You can work with Claude itself to build skills and plugins. Describe what you want the workflow to do, what your firm’s processes look like, what outputs you need, and Claude will help you draft, test, and refine the skill. Anthropic has also open-sourced eleven starter plugins on GitHub (https://github.com/anthropics/knowledge-work-plugins/tree/main/legal) that you can customize. The barrier to entry here is remarkably low. If you can write a clear memo, you can write a skill.
Now Let’s Talk About Claude Cowork
Claude Cowork is Anthropic’s desktop tool that takes Claude out of the chat window and puts it to work on your actual files. You point it at a folder on your computer, give it a task, and it plans, executes, and iterates through multi-step workflows on its own. It can read documents, create new files, delete files, organize folders, and coordinate multiple workstreams. If you’ve used Claude Code (the developer-facing terminal tool), Cowork is the same engine with a friendlier interface. It’s essentially Claude Code for laptop professionals who don’t want to learn code. It is currently only available in iOS but Claude recently launched the Claude app for PC, so we’re hopeful to see them in that soon.
What makes Cowork interesting for legal professionals is that it doesn’t just respond to one prompt at a time. You set a goal, and Claude works through it like a (very fast, very thorough) junior associate. It asks for clarification when it needs it, saves outputs directly to your file system, and loops you in on its progress. The plugin system takes this further: install a legal plugin, and Cowork immediately knows your firm’s preferred tone, your risk tolerances, your clause library, and your review workflow, what files to reference as template, etc. It’s a configurable work environment; no more cutting and pasting required.
One more thing worth noting, because I think it says something profound about where we are: Cowork was built entirely by Claude Code in approximately 10 days. An AI coding agent built its own non-technical sibling, and that sibling then crashed the legal tech market. 2026 is going to be wild.
Practical Examples: Skills for Lawyers and Legal Academics
Enough theory; let’s build something. Below are four practical examples, two for practicing attorneys and two for legal academics, showing how you could combine Claude Skills and Cowork to create custom workflows. Each example includes the actual markdown you’d use to create the skill. Markdown is still human-readable, it just gives useful formatting to the text for an LLM. Basically, any complex prompting I do these days is in markdown.
For Lawyers
- Client Intake and Conflict Check Workflow
This skill automates the initial client intake process: reviewing incoming matter information, running a basic conflict check against your existing client list, and generating a standardized intake memo.
—
name: client-intake-workflow
description: Automates client intake processing, conflict checking, and memo generation for new matters
—
# Client Intake and Conflict Check Workflow
## Purpose
Process new client intake forms, check for potential conflicts against the firm’s existing client and matter database, and generate a standardized intake memo for partner review.## Trigger
Use this skill when processing new client intake information, when a new matter is opened, or when asked to run a conflict check.## Intake Processing Steps
**Step 1: Extract Key Information**
From the intake form or email, extract and organize:
– Prospective client name (individual and entity)
– All related parties (opposing parties, co-defendants, subsidiaries, affiliates)
– Nature of the matter (practice area, dispute type)
– Relevant dates and deadlines
– Referring source
– Estimated matter value or scope**Step 2: Conflict Check**
Search the firm’s client/matter list for potential conflicts:
– Direct name matches (exact and fuzzy matching)
– Entity relationships (parent companies, subsidiaries, DBAs)
– Opposing party matches across all open and closed matters
– Flag any matter where a related party appears on both sidesConflict Classification:
– **CLEAR**: No matches found across any category
– **POTENTIAL**: Fuzzy matches or related entity connections that require human review
– **CONFLICT**: Direct match on adverse parties in existing or prior matters, requires partner review before proceeding**Step 3: Generate Intake Memo**
Produce a memo in the following format:*Header:* New Matter Intake Memo, [Date], [Matter Name]
*Sections:*
1. Client Information (name, contact, entity type)
2. Matter Summary (2-3 sentence description)
3. Conflict Check Results (classification + details)
4. Recommended Next Steps
5. Assigned Attorney / Responsible Partner (leave blank for assignment)## Output
Save the memo as `[ClientName]-intake-memo-[date].docx` in the current working folder.## Important Notes
– Always flag potential conflicts for human review. Never clear a conflict automatically.
– This workflow assists with intake processing. All conflict determinations must be reviewed and approved by a licensed attorney before the firm accepts representation.
– If the intake information is incomplete, list the missing fields and draft a follow-up request email to the prospective client.
How to use it: Drop your intake emails or forms into a Cowork folder with this skill installed. Tell Cowork: “Process the new intake from the Johnson matter.” It reads the documents, runs the conflict check against your client list (which you’ve also placed in the folder or connected via MCP), and generates a clean intake memo for partner review. What used to take 30-45 minutes of admin time now takes about 2 minutes.
- Contract Review and Risk Flagging
This skill creates a systematic contract review process aligned with your firm or organization’s specific risk tolerances and playbook.
—
name: contract-risk-review
description: Reviews contracts against firm playbook, flags risk levels, and generates a redline summary with recommended changes
—# Contract Review and Risk Flagging
## Purpose
Review incoming contracts against the firm’s standard positions, flag deviations by risk level, and produce a structured summary with recommended modifications.## Trigger
Use this skill when asked to review a contract, NDA, MSA, vendor agreement, or any document requiring clause-by-clause risk analysis.## Firm Playbook Defaults
(Customize these to match your firm or client’s standard positions)**Acceptable Positions (Green)**
– Mutual confidentiality obligations
– 12-month non-solicitation of employees
– Standard reps and warranties
– Limitation of liability capped at contract value
– 30-day cure period for material breach
– Governing law: [your preferred jurisdiction]**Requires Negotiation (Yellow)**
– Unilateral confidentiality obligations
– Non-compete broader than 12 months or geographic scope
– Indemnification without cap or carve-outs
– Auto-renewal clauses exceeding 12 months
– Assignment without consent
– Audit rights without reasonable notice**Unacceptable / Escalate Immediately (Red)**
– Unlimited liability
– Waiver of jury trial (unless pre-approved by client)
– Unilateral termination for convenience without notice period
– IP assignment of pre-existing work product
– Non-mutual arbitration clauses
– Choice of law in jurisdictions where firm/client has no presence and did not agree to## Review Process
**Step 1: Document Identification**
Identify the contract type, parties, effective date, and term length.**Step 2: Clause-by-Clause Analysis**
Review each material clause and classify as:
– **GREEN**: Aligns with firm standard positions
– **YELLOW**: Deviates from standard but is negotiable
– **RED**: Unacceptable, requires immediate escalationFor each Yellow and Red clause: quote the relevant language, explain the risk, and provide recommended alternative language.
**Step 3: Cross-Reference Check**
Flag any clauses that interact with or contradict each other (e.g., an uncapped indemnity alongside a limitation of liability creates internal inconsistency).**Step 4: Generate Review Summary**
Produce a summary document with:
1. Contract overview (parties, type, term, value)
2. Risk summary (count of Green/Yellow/Red flags)
3. Detailed findings table (clause, classification, issue, recommendation)
4. Priority action items (Red flags listed first)
5. Recommended next steps## Output
Save as `[ContractName]-review-[date].docx` in the current working folder.## Important Notes
– This review is for attorney use as a first-pass tool. All recommendations must be reviewed by the responsible attorney before being communicated to opposing counsel or the client.
– Flag but do not attempt to resolve ambiguous provisions. Note them for attorney review.
– If the contract references incorporated documents (exhibits, schedules, prior agreements) that are not provided, note them as missing and flag that the review is incomplete without them.
How to use it: Drop a contract PDF into your Cowork folder and say: “Review this MSA against our playbook.” Cowork reads the entire document, classifies every material clause, flags the risks, and produces a structured review memo with recommended redline language. Customize the playbook section to match your firm’s actual standard positions, and you’ve just built a bespoke contract review system that would cost five or six figures from a vendor.
- Legal Writing Review and Structural Editing
Anyone who writes for a living knows the pain of staring at your own draft and not being able to see its problems anymore. This skill gives you a structured second set of eyes. It reviews legal writing (briefs, memos, articles, motions, scholarly papers) for structural coherence, argument flow, citation consistency, and clarity. For practitioners, it catches the weak transitions and unsupported assertions before opposing counsel does. For academics, it flags the structural issues a law review editor would catch, before submission.
—
name: legal-writing-review
description: Reviews legal writing for structural coherence, argument flow, citation issues, and clarity. Works across briefs, memos, motions, articles, and scholarly papers.
—# Legal Writing Review and Structural Editing
## Purpose
Provide a structured editorial review of legal writing, focusing on organization, argument flow, citation usage, and clarity. This is not a grammar checker. It evaluates whether the document makes its case effectively and identifies structural weaknesses.## Trigger
Use this skill when asked to review, edit, or provide feedback on legal writing including briefs, memos, motions, law review articles, blog posts, op-eds, or any professional legal document.## Review Process
**Step 1: Document Assessment**
Identify and log:
– Document type (brief, memo, article, motion, etc.)
– Intended audience (court, client, law review, general public, etc.)
– Core thesis or argument (state it in one sentence)
– Total length and section count
– Citation style in use (Bluebook, ALWD, informal)**Step 2: Structural Analysis**
Evaluate the document’s architecture:
– Does the introduction clearly frame the issue and signal the conclusion?
– Does each section advance the core argument, or are there sections that drift or repeat?
– Is the ordering of sections logical? Would rearranging strengthen the argument?
– Are transitions between sections and paragraphs clear and purposeful?
– Does the conclusion do more than summarize? Does it land the argument?
– Flag any section longer than 3 pages that lacks internal subheadings**Step 3: Argument Flow Analysis**
For each major argument or section:
– State the claim being made
– Identify the supporting evidence or authority cited
– Flag any claims that lack supporting authority
– Flag any authority cited but not meaningfully analyzed (dropped citations)
– Identify logical gaps: places where the reader must make an inferential leap the author hasn’t bridged
– Note any counterarguments that are raised but insufficiently addressed**Step 4: Citation Review**
Review citations for:
– Consistency of format throughout the document
– Signals: are citation signals (see, see also, cf., but see) used correctly and consistently?
– String citations: flag any string of 4+ citations without explanatory parentheticals
– Recency: flag any authority older than 10 years that is presented as current law without acknowledging its age
– Missing citations: flag any factual or legal assertion that appears to need a citation but lacks one
– Note: Do not verify that citations are real. Flag formatting issues and missing citations only.**Step 5: Clarity and Readability**
Evaluate sentence-level effectiveness:
– Flag sentences longer than 40 words that could be split without losing meaning
– Flag passive voice where active would be stronger (especially in argument sections)
– Flag nominalizations that weaken the prose (e.g., “made a determination” instead of “determined”)
– Flag unnecessary throat-clearing (e.g., “It is important to note that…” or “It should be emphasized that…”)
– Flag jargon or legalese that could be simplified without losing precision**Step 6: Generate Review Memo**
Produce a review memo with:
1. **Document Summary** (type, length, core thesis)
2. **Overall Assessment** (2-3 sentence evaluation of the document’s effectiveness)
3. **Structural Recommendations** (reordering, consolidation, expansion, or cuts)
4. **Argument Gaps** (specific claims that need stronger support or counterargument treatment)
5. **Citation Issues** (list with page/paragraph references)
6. **Line-Level Suggestions** (top 10 most impactful sentence-level improvements, not an exhaustive grammar check)
7. **Strengths** (what the document does well, identify 2-3 specific strengths)## Output
Save as `[DocumentName]-review-[date].docx` in the current working folder.## Important Notes
– This is editorial feedback, not legal advice. Do not evaluate the legal merits of arguments.
– Preserve the author’s voice. Suggest structural changes, not rewrites. The goal is to make the author’s argument stronger, not to rewrite it in a different voice.
– Be specific. “This section is unclear” is unhelpful. “This section claims X but only cites Y, which supports a different proposition” is useful.
– For academic writing, note whether the article’s contribution to the literature is clearly stated in the introduction.
– Prioritize feedback by impact. Lead with the changes that would most improve the document.
How to use it: Drop a draft brief, memo, or article into a Cowork folder and say: “Run a full writing review on this document.” Cowork reads the entire piece, maps the argument structure, checks citation consistency, identifies logical gaps, and produces a structured review memo prioritized by impact. This is the kind of detailed structural feedback that associates get from senior partners and that scholars get from peer reviewers, except it’s available at 2 AM on a Sunday when you’re trying to finalize a filing or hit a submission deadline. I’ve found it particularly useful for catching the “forest for the trees” problems that are invisible after your fifth read of your own draft.
More of an Academic Use-Case
- Research Literature Review and Gap Analysis
This skill processes a collection of academic articles and produces a structured literature review with identified gaps and potential research directions.
—
name: lit-review-gap-analysis
description: Processes a collection of academic articles, generates a structured literature review organized by theme, and identifies research gaps and potential directions
—# Research Literature Review and Gap Analysis
## Purpose
Read and analyze a collection of academic articles, organize findings thematically, identify points of agreement and disagreement in the literature, and surface gaps where further research is needed.## Trigger
Use this skill when asked to conduct a literature review, organize research sources, identify research gaps, or prepare a “state of the literature” summary for a specific topic.## Process
**Step 1: Source Inventory**
For each article in the provided folder, extract:
– Full citation (authors, title, journal, year, volume, pages)
– Research question or thesis
– Methodology (empirical, doctrinal, theoretical, mixed-methods)
– Key findings or arguments (3-5 bullet points)
– Limitations noted by the authors
– Relevant data sources or datasets referenced**Step 2: Thematic Coding**
Group the articles by theme. Identify:
– Primary themes (topics addressed by 3+ articles)
– Secondary themes (topics addressed by 1-2 articles)
– Cross-cutting themes (issues that appear across multiple primary themes)For each theme, note:
– Points of consensus across sources
– Points of disagreement or tension
– Methodological approaches used
– Chronological development of the theme (how thinking has evolved)**Step 3: Gap Analysis**
Identify gaps in the literature:
– **Empirical gaps**: Questions that lack data or where existing data is outdated (older than 3 years for fast-moving fields like legal tech)
– **Methodological gaps**: Topics studied only through one lens (e.g., only surveys, no backend analytics)
– **Population gaps**: Groups or practice areas underrepresented in the research
– **Theoretical gaps**: Frameworks that haven’t been applied to this domain but could be productive**Step 4: Generate Literature Review Document**
Structure the output as:
1. **Introduction** (scope of review, number of sources, date range covered)
2. **Methodology** (how sources were selected and analyzed)
3. **Thematic Analysis** (organized by primary theme, with sub-sections for secondary themes)
4. **State of the Debate** (key areas of agreement and disagreement)
5. **Research Gaps and Opportunities** (organized by gap type)
6. **Suggested Research Directions** (3-5 specific proposals with brief justifications)
7. **Source Table** (complete citation list with thematic tags)## Output
Save as `[TopicName]-lit-review-[date].docx`## Important Notes
– This tool organizes and synthesizes. It does not generate or fabricate sources. Every claim must be traceable to a specific article in the provided collection.
– If the collection appears to be missing key voices or perspectives on a topic, note this as a limitation.
– For legal scholarship, note whether sources are law review articles, practitioner publications, empirical studies, or policy papers, as the methodological weight differs.
– When identifying gaps, distinguish between “nobody has studied this” and “this has been studied but with limited data or outdated methods.”
How to use it: Dump your collected PDFs and articles into a Cowork folder and say: “Run a literature review and gap analysis on these sources.” Cowork reads every article, codes them thematically, identifies where the scholarship agrees and disagrees, and surfaces the specific gaps where your next paper could live. For anyone who has stared at a pile of 40 PDFs trying to figure out where the conversation is and where it isn’t, this is a genuine time-saver. I use a version of this workflow for my own research on AI diffusion in legal services, and it’s transformed how I approach the early stages of a new article.
What This Actually Means
Look, I want to be clear about something: these skills don’t replace lawyers, and they don’t replace legal scholars. What they do is eliminate a staggering amount of tedious, repetitive work that currently eats up hours of professional time. The contract review skill doesn’t decide whether to accept the indemnification clause. It flags the clause, tells you why it deviates from your playbook, and suggests alternative language. You still make the call.
But the structural implications are real. When a 250-line markdown file can replicate functionality that legal tech companies have spent years building and charging thousands of dollars for, the competitive landscape shifts. That’s what the market reacted to. Not the current state of the technology, but the trajectory. If this is what a $20/month subscription can do today, what happens in six months? A year?
Start building skills. Start with something simple, a intake checklist, a grading rubric generator, a research organization tool. Work with Claude to refine it. then get more ambitious. The firms and law schools that build institutional knowledge into reusable, customizable AI workflows will have a significant competitive advantage. The ones that wait for vendors to sell them a solution will pay more and get less.
The tools are here. They’re $20 a month. And, apparently, they can crash a stock market.
Sean A. Harrington is Director of the AI & Legal Tech Studio at Arizona State University College of Law, where he teaches AI and the Practice of Law among other AI initiatives. He speaks nationally on AI implementation for attorneys and legal educators.