Personal Claude setups are quick to build. The shared knowledge layer underneath them is what turns AI from an individual tool into a team capability.
AI adoption among legal professionals is accelerating, but most of that adoption still lives in chats and on individual laptops. For example, two lawyers on our legal team ran the same Claude skill on a standard contract clause and got back two different versions of it. The model was fine. But the personal Claude setups underneath were not aligned, and there was no shared knowledge layer holding them together. The team rebuilt their setup around a single Claude Cowork project that mirrors our Google Drive structure (shared playbooks, a custom skills library, instruction files, a folder protocol, a naming convention) to scale Claude’s legal AI capabilities beyond individual use.
01 Where Personal Setups Stopped WorkingLast month, a colleague and I were drafting similar vendor agreements. We both ran the same Claude skill on the indemnification clause and got back two different versions: mine used our current standard, his used one we had retired a few weeks earlier. Neither of us caught minor differences until peer review.
Our personal Claude setups were good. What we did not have was a proper knowledge layer shared underneath them.
The vendor agreement incident was not isolated. Once we started looking, we saw the same pattern across other work. The same skill, run by two different people, produced different formatting, different tone, and sometimes different standard terms in a contract clause. Different formatting was a nuisance we could deal with. Different contract terms were not. In legal, outputs must be interchangeable, built on the same agreed baseline, and consistent enough that a colleague can review them without second-guessing where the variation came from, or why.
The problem was that each person’s AI was drawing from a different context. Everyone had their own instructions, their own templates, and even their own assumptions about how the team operates.
This is not unique to Provectus’ legal team. Thomson Reuters’ 2026 AI in Professional Services Report found that generative AI adoption among legal professionals more than doubled between 2025 and 2026, from 31% to 69%. In the meantime, Adobe’s 2026 AI and Digital Trends Report found that 57% of organizations say AI is changing workflows faster than employees can adapt. The headline numbers describe individual adoption. They do not describe, however, whether the same AI works the same way across a team.
Anthropic understands the problem, too. In May 2026, the company launched Claude For Legal with role-specific plugins and more than twenty legal integrations; firms like Freshfields rolled it out fast enough to see roughly 500% growth in Claude usage within six weeks. More lawyers using more AI in more workflows means more places where outputs can drift apart, unless a robust knowledge layer underneath holds them together.
02 The Shared Workspace SetupWe already had a shared Google Drive folder with templates, playbooks, and reference material. The goal was to build the equivalent of this knowledge layer for Claude: a shared space everyone works from, with restricted access and a clear structure.
We set up a single Claude Cowork project that mirrors our existing folder structure. Inside it we store:
The last three items tend to get skipped in most setups. They are also the things that keep a shared space from reverting to the same fragmentation it was built to solve. A shared folder without structure and ownership rules gradually becomes just another pile of documents.
03 What Changed After SetupThough we did not run a controlled measurement, we can report that four things have changed in our day-to-day work.
Two lawyers reviewing the same contract type no longer produce different indemnification or limitation-of-liability terms by accident. They diverge only when one of them deliberately edits the standard. The baseline lives in one place now, so it can only be one thing.
A new lawyer running our contract review skill gets output that looks like the rest of the team’s, because the skill itself carries the playbook and the standards. They do not have to rebuild that context inside a personal setup before they can contribute anything useful.
When one of us finds a better way to structure a clause, or a better prompt for a recurring task, the update goes into the shared skill once. The next person to run it picks it up. Previously, that improvement would have lived in one person’s setup until they happened to mention it in a Slack thread.
We no longer spend time reconciling which copy of a template is authoritative, because there is one copy and Claude pulls from it.
Note: Axiom’s 2025 in-house legal AI report found that only 21% of in-house legal teams have reached AI maturity, despite widespread adoption. A shared workspace can help close the gap between using AI individually and operating AI as a team.
04 The Meta-SkillOnce skills became shared infrastructure, something we had not expected became possible: we could build skills that improve other skills.
For example, we developed one specifically for creating and refining the rest of our library. It reads an existing skill’s instructions, tests them against sample inputs, suggests changes, and tracks version history as the system evolves. If skills live on individual laptops, a tool like this is pointless. Once they live in a shared workspace, every improvement the meta-tool suggests reaches the whole team.
The team builds on a skill in the shared workspace. A skill on one person’s laptop gets used and forgotten.
05 What This Means for Legal TeamsMost legal teams in 2026 are already using AI. The harder problem now is whether the AI running across the team pulls from the same source of truth. Without alignment, the productivity gains AI was supposed to deliver get eaten by coordination overhead and rework.
Three questions worth asking your team first:
If you cannot answer the first one, the other two answer themselves.
Building the shared knowledge layer underneath a team is the partner job. It is what Provectus does for legal, finance, operations, and customer-facing teams across our client base.
The work depends on where the team starts. Some need the shared workspace built from scratch: the skills library, the playbook digitization, the folder protocol, the naming convention. Some have a workspace but lack the protocols that keep it from sliding back into fragmentation. Some have both, and need the training that turns “we have a shared workspace” into “people actually use it.” Fifteen years of bringing AI to production sits behind this work, augmented by our partnership with Anthropic. Industry knowledge sits alongside technical depth, because the workflows that matter most to any team depend on understanding the work itself, not just the tooling.
If you are interested in exploring how a shared AI workspace could look inside your legal or operational team, reach out to us and we will start the engagement.