Skip to content
ClaudeClaude

Claude Cowork for legal teams

Legal teams are being asked to do more with less. A big part of the job is coming up to speed fast: a question lands about a feature that launched months ago, and the context you had when you wrote the original memo is gone. This video shows how an in-house product lawyer uses Claude Cowork to get back up to speed in minutes. Claude pulls the past review and the latest threads, then hands back a short brief that points to exactly what changed. Every claim traces back to a source, so you can verify before your name goes on the reply. Claude Cowork is Claude for knowledge work. It connects to the tools you already use, runs multi-step tasks with your approval, and runs on skills that fit your workflow. Try Claude Cowork: claude.com/cowork

May 18, 20262mWatch on YouTube ↗

CHAPTERS

  1. 0:09 – 0:39

    Rapid legal question arrives with no context: the daily in-house reality

    Mark, an in-house product lawyer, gets a last-minute Slack question about a feature he previously reviewed. With only minutes before a meeting and the original context long gone, he frames the typical time sink of reconstructing past analysis.

    • Incoming Slack question about an older product launch
    • Time pressure before the next meeting
    • Lost context from the original memo and review
    • Old workflow would require rereading prior work and tracking changes
  2. 0:39 – 0:49

    Scheduling Claude Cowork as a morning “chief of staff” briefing

    He describes setting up Claude Cowork to run a scheduled task each morning. The goal is a daily memo that prioritizes what’s new, urgent, and due, so he can start from an organized snapshot instead of digging through tools.

    • Automated morning task run in Claude Cowork
    • Daily memo of what’s on the lawyer’s plate
    • Prioritization by urgency and due dates
    • Reduces time spent triaging across systems
  3. 0:49 – 0:59

    Connecting email and triaging work items by due date

    With Gmail integrated, Claude Cowork surfaces a short, prioritized list of items due today. Mark selects the relevant follow-up item tied to the product manager’s request.

    • Gmail inbox connected for unified context
    • Items grouped/sorted by due today vs. new vs. routine
    • Quick selection of the relevant follow-up thread
    • Less manual scanning of inbox and messages
  4. 0:59 – 1:09

    Using the /brief skill to get up to speed on a product change fast

    Mark relies on a key workflow, “/brief,” to rapidly reorient on a specific product and the precise change in question. This is positioned as the primary accelerant for regaining context without rereading everything.

    • /brief as the core skill for rapid context-building
    • Focused on a named product and specific change
    • Designed for fast ramp-up on a narrow legal question
    • Avoids re-reading entire prior memos to find relevant sections
  5. 1:09 – 1:19

    Open, customizable legal plugin built from Anthropic’s internal playbooks

    He explains the /brief plugin was built based on how Anthropic’s legal department operates and is offered as an open protocol. Teams can tailor it to their own templates and playbooks rather than adopting a one-size-fits-all workflow.

    • Plugin reflects real internal legal workflows
    • Open protocol that others can adapt
    • Customizable to company-specific playbooks and templates
    • Encourages standardization without rigidity
  6. 1:19 – 1:29

    Minimal prompting: Claude already knows the review file locations and structure

    Mark’s prompt stays short because the system is configured to understand where review artifacts live and how templates are structured. That setup lets Claude do the heavy lifting of retrieval and organization.

    • Short prompt enabled by structured setup
    • Preconfigured knowledge of folder locations and templates
    • Workflow tuned to legal review artifacts
    • Less time specifying where to find materials
  7. 1:29 – 1:39

    Automated retrieval + synthesis across Slack, Gmail, and prior reviews

    Claude pulls the relevant review materials, reads the new Slack and email threads, and synthesizes a brief in the format Mark prefers. The output highlights past conclusions, what changed, and where the prior analysis is impacted.

    • Pulls prior review from the correct folder
    • Reads current Slack and Gmail threads for new context
    • Summarizes: prior conclusion, current change, impacted analysis
    • Returns information in a lawyer-friendly structure
  8. 1:39 – 1:49

    Precision over volume: pointing to the exact paragraphs that matter

    Instead of forcing Mark to scan a long memo, the brief directs him to the specific sections that answer the current question. This turns a document search problem into a targeted verification step.

    • Avoids rereading a 40-page memo for a few paragraphs
    • Brief identifies the precise relevant sections
    • Direct linking into earlier review content
    • Speeds up comprehension and response preparation
  9. 1:49 – 1:59

    Human-in-the-loop verification: trust but verify

    Mark checks the quoted paragraph verbatim because he remains accountable for the final guidance. He emphasizes a deliberate human-in-the-loop approach where AI accelerates retrieval and drafting, but legal judgment stays with the lawyer.

    • Verification before sending guidance
    • Accountability: “my name goes on the reply”
    • Human-in-the-loop as a core control
    • AI used to reduce busywork, not replace judgment
  10. 1:59 – 2:10

    From busywork to strategic legal thinking

    By removing document hunting and context reconstruction, Mark frees time for higher-value analysis. He frames this shift as focusing on the work lawyers trained for rather than “work about the work.”

    • Less time spent searching and assembling documents
    • More time for strategic legal reasoning
    • Reduces administrative overhead of legal support
    • Aligns day-to-day work with professional training
  11. 2:10 – 2:25

    Drafting the response and getting explicit approval before sending

    Once confident, Mark asks Claude to draft a reply to the product team. Claude requests approval, reinforcing the review-and-send workflow and keeping the lawyer in control of external communication.

    • Claude drafts a response to the product team
    • Approval step before the message is sent
    • Maintains lawyer control over final language
    • Accelerates responsiveness to fast-moving teams
  12. 2:25 – 2:40

    Closing the loop in Jira and building shared institutional knowledge

    Mark ties the interaction into Jira ticketing so the resolution is recorded for the team. Over time, these captured decisions form a searchable corpus that reduces information silos and improves future turnaround.

    • Integrates with Jira to close out tickets
    • Creates durable context for future questions
    • Builds a knowledge corpus for legal (and possibly broader) access
    • Reduces reliance on individual memory and siloed docs
  13. 2:40 – 2:59

    Outcome: faster responsiveness and “do more with less” in legal

    He ends with the practical impact: he can enter his next meeting confident and responsive. Mark connects this workflow to the broader pressure on legal teams to scale output with limited resources while adopting AI responsibly.

    • Quick turnaround on product team requests
    • Confidence going into subsequent meetings
    • Supports the mandate to “do more with less”
    • AI adoption framed as empowering and practical

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.