Claude for Legal artifacts: what they are, how to use them
2026-10-28 · 4 min read
Claude for Legal artifacts are ready-made interactive tools, such as trackers, tables and dashboards, that a legal team opens inside Claude, tries on a sample, then points at its own documents and connectors. Anthropic lists the gallery under Claude Enterprise, and you browse it at claude.ai/artifacts?templates=legal.
The launch came from the @claudeai account today, announcing "a gallery of ready-made artifacts for contract review, regulatory horizon scans, cite checks" and more. Within the hour, @VaibhavSisinty summed it up as Anthropic turning Claude for Legal into "a shelf of ready-made workflows." On the Chinese-language side, @dotey wrote a longer breakdown and pointed to the detail that matters most for lawyers: every conclusion an artifact gives links back to the source text, so it can be checked.
What the six Anthropic artifacts do
These come from Anthropic's announcement. Each one opens on a worked example, so you see the output before you load anything real.
| Artifact | What it does |
|---|---|
| Cite check | Compares a brief or memo against its sources and flags misquotes, wrong page numbers and citations that don't support the claim |
| Jurisdiction survey matrix | Puts one legal topic, such as breach-notice rules, side by side across states or countries, linked to the law |
| Matter board | Turns engagement letters, emails or spreadsheets into a board of open matters with owner, next step and next deadline |
| Playbook deviation board | Checks contracts against your standard terms, flags clauses that differ and routes them to whoever approves exceptions |
| Regulatory horizon scan | Lays upcoming regulations on a timeline for your business and locations, ranked by impact, with source links |
| Tabular review | Builds a table of terms you ask about (severance, non-competes, change of control) across a stack of contracts, linked to the exact language |
Anthropic describes the division of labor plainly: the lawyer supplies the documents, Claude does the first pass, the lawyer makes the call.
Partner and access-to-justice artifacts
Four outside builders shipped artifacts for the launch. Goodwin built an AI Governance Benchmark, Foley & Lardner an Obligations Checker, and PwC a Subpoena Response Tool. Harvey's entry, Harvey Intake, gives employees one place in Claude to ask legal questions. It connects to Harvey through Harvey's MCP server, answers routine questions from Legal's playbooks and sends anything that needs judgment to a lawyer. Harvey doesn't spell out eligibility, but expect to need a Harvey account for that one. Anthropic says Legora and Orrick artifacts are next.
There are also nine access-to-justice artifacts. Five come from nonprofits that received Claude credits, including a rent ledger reviewer and a record-sealing screener from Rising for Justice (both based on D.C. law) and an immigration papers reader from Mobile Pathways.
How to try it
- Open the gallery at
claude.ai/artifacts?templates=legal, or go to the Artifacts tab in the Claude sidebar, which Anthropic's help center says doubles as your template gallery. - Make sure "Cloud code execution and file creation" is turned on in settings. The help center lists it as a requirement for artifacts.
- Open one, say tabular review, and look at the sample output before you upload anything.
- Ask Claude to tailor it. A prompt like "Swap in the 14 vendor agreements in this folder and add columns for auto-renewal date and termination notice period" is the kind of edit the announcement describes.
- Connect sources instead of uploading. Anthropic says it has added more than 60 legal connectors since May, including Bloomberg Law, Filevine and Legora.
- Share it. Artifacts start private, and Anthropic says the sharing settings are built to respect ethical walls between teams.
If you have not used artifact templates before, our walkthrough of Claude Dashboards covers the same Output menu and Artifacts tab.
My read
Lawyers could already ask Claude to compare contracts. The gallery hands them a working first draft of the tool, source links included, so a firm doesn't start from a blank prompt and an empty layout. Cite check and tabular review fix problems every firm has. For in-house teams, I'd bet the playbook board saves the most hours.
The catch is access. This sits under Claude Enterprise, and Anthropic routes rollout through sales and implementation partners. Small firms on Team or Pro should not expect the legal gallery to show up yet. Compare that with OpenAI, which gave Astra for Law to big firms first. Both labs are starting at the top of the market.
What this means for a small business
The pattern is worth copying even if you never touch Enterprise. Give a repeated review task one fixed template, make every output link back to its source, and have a person sign off. Any office with a review queue can run it that way, legal or not. If your firm is still picking its practice software, our Clio vs MyCase comparison is a good place to start. If you want help finding which of your own reviews could run this way, New Face Design offers a free process audit.