# Build Your Own AI Legal Team in Claude > Source: > Published: 2026-10-31 21:32:28+00:00 [“A few hundreds of dollars per seat, per month.”](https://www.fastcompany.com/91429116/harvey-openai-and-the-race-to-use-ai-to-revolutionize-big-law) That’s what a Harvey spokesperson told Fast Company a bespoke seat costs, and large customers buy at least 100 seats for at least a year. [Claude Pro costs $20 a month](https://claude.com/pricing). **By our reading, that’s at least 90% less per seat**: we take “a few hundred” as $200 or more, and $20 is 10% of $200. Harvey is built for law firms with hundreds of lawyers. A seed-stage startup has zero lawyers and a folder of PDFs someone has to read before Friday. The other price tag is outside counsel. Clio puts the average US lawyer at [$362 an hour, and business formation and compliance work at $400](https://www.clio.com/resources/legal-trends/compare-lawyer-rates/). **One of those hours pays for 2 years of Claude Pro** on the [$200 annual plan](https://claude.com/pricing). Anthropic shipped [legal plugins for Claude Cowork](https://techcrunch.com/2026/01/30/anthropic-brings-agentic-plugins-to-cowork/) on January 30, 2026, with [contract review and NDA triage](https://github.com/anthropics/knowledge-work-plugins/tree/main/legal) among them. On February 3, Thomson Reuters shares closed down about 16%, per [The Wall Street Journal](https://www.wsj.com/livecoverage/stock-market-today-dow-sp-500-nasdaq-02-03-2026/card/C97KvtPTsT9H3joFwS7C). Cheap only works with guardrails. In [Vals AI’s legal benchmark](https://www.vals.ai/industry-reports/vlair-2-27-25), Harvey scored **94.8% on document Q&A against 70.1% for the lawyer baseline**, and lawyers still won redlining, 79.7% to 65.0% for the best tool. Reading is where AI already wins. Judgment and drafting the final word still belong to a human. A [Stanford study](https://law.stanford.edu/wp-content/uploads/2024/05/Legal_RAG_Hallucinations.pdf) found that even purpose-built legal research tools hallucinated on more than 17% of queries. Damien Charlotin’s database now tracks [2,241 court cases with AI hallucinations](https://www.damiencharlotin.com/hallucinations/), and 884 of them involve lawyers. So the rule for a founder’s AI legal team is simple: **AI reads everything and quotes every line it relies on, then a lawyer decides the few things that can hurt you.** Try it on the next NDA in your inbox. Copy & paste this: ``` You are reviewing an NDA for [COMPANY]. We are the [DISCLOSER / RECIPIENT / MUTUAL] party. 1. Quote every clause you rely on, word for word, with its section number. If you cannot quote it, drop the point. 2. Check: mutual or one-way, definition of confidential information, standard exclusions, term and survival, residuals, non-solicit, non-compete, IP or feedback assignment, remedies, governing law. 3. Give one verdict: SIGN, NEGOTIATE or ESCALATE (a lawyer reads it first). 4. For every NEGOTIATE item, write: Replace "[exact quote]" with "[new text]". 5. Cite no law from memory. Turn any legal question into a question for my lawyer. End with: "Prepared by AI for review. This is not legal advice." ``` That prompt handles one NDA. A startup signs NDAs, SAFEs, offer letters, customer MSAs and DPAs, files Delaware taxes and keeps a cap table that has to match all of it. Each of those documents needs its own checks and its own red flags. Most founders paste each one into a chat and trust the confident answer that comes back. One invented clause in a customer contract or one missed 83(b) deadline costs more than a year of any AI plan. So I built the kit I wanted for that moment. On the sample NDA inside it, the verifier script caught **1 invented quote out of 4 findings and 1 law citation with no source**, before anything reached a lawyer. Paid subscribers get the full kit today: 1. **10 ready-to-install Claude Skills** , one for each legal moment a founder hits plus the checks that run on every task 2. **An “opposing counsel” agent** that attacks your redlines before the other side does 3. **3 scripts** : a quote verifier, a cap table checker and a compliance calendar that exports to Google Calendar 4. **A playbook template and an issue-list schema** , plus a 4-level escalation matrix and a 1-day build plan A few that pair well with this one: [The Claude Prompts That Close Rounds](https://www.the-ai-corner.com/p/the-claude-prompts-that-close-rounds) (in [Business & Investing](https://www.the-ai-corner.com/t/business-and-investing)), [The Claude Code System That Replaces a 5-Person Team](https://www.the-ai-corner.com/p/the-claude-code-system-that-replaces) (in [Claude & Anthropic](https://www.the-ai-corner.com/t/claude-and-anthropic)), [The Agent Workforce Kit](https://www.the-ai-corner.com/p/the-agent-workforce-kit-anthropics) (in [AI Agents](https://www.the-ai-corner.com/t/ai-agents)) and [Build a One-Person Business With Claude](https://www.the-ai-corner.com/p/build-one-person-business-with-claude) (in [Resources](https://www.the-ai-corner.com/t/resources)). **Get 50% off for 1 year.** Premium also opens the full library: the [Resources](https://www.the-ai-corner.com/t/resources) hub, every [Claude & Anthropic](https://www.the-ai-corner.com/t/claude-and-anthropic) playbook, the [AI Agents](https://www.the-ai-corner.com/t/ai-agents) builds and the [Business & Investing](https://www.the-ai-corner.com/t/business-and-investing) deep dives.