Anthropic adds legal artifacts to Claude Enterprise, with cite checks

Anthropic said on 28 October 2026 that Claude for Legal, launched in May, now includes a gallery of ready-to-use artifacts: interactive trackers, boards and tables that a legal team can adapt with its own data and share across the business. The company says the most common requests since May were for simpler setup and easier ways to work with colleagues. It lists the product as part of Claude Enterprise.

What is in the gallery

Anthropic's own six artifacts are described as follows.

• Cite check compares a legal document with its sources and flags quotes that do not match, wrong page numbers and sources that do not support the claim.
• Jurisdiction survey matrix compares a topic, such as data breach notice rules, across states or countries in one table with links to the underlying law.
• Matter board turns engagement letters, emails or spreadsheets into one board of open matters with owner, next step and next deadline.
• Playbook deviation board compares contracts with a company's standard terms, flags each clause that differs and sends it to the person who approves exceptions.
• Regulatory horizon scan maps upcoming regulations on a timeline based on where the company operates, ranked by impact with source links.
• Tabular review builds a table from a set of contracts for terms such as severance or non-competes, with links to the exact language.

Each artifact opens with an example so a legal professional can see the result before adding their own documents. Anthropic describes the workflow as the user bringing documents and sources, Claude completing the first pass, and a lawyer applying judgment to the results.

Partners and connectors

Four artifacts come from outside Anthropic: an AI Governance Benchmark from Goodwin, Harvey Intake from Harvey, an Obligations Checker from Foley & Lardner and a Subpoena Response Tool from PwC. Anthropic says more are coming from Legora and Orrick. It also reports more than 60 legal connectors added since May, so playbooks and research sources can be pulled in without manual uploads.

The Decoder, reporting on the launch, adds that the model handles an initial verification and links each statement to its original source before a lawyer reviews it. It repeats Anthropic's claim that built-in settings enforce data protection rules and client-required information barriers between teams.

What the announcement leaves out

Anthropic's page gives no error rates for the cite check, no pricing for the artifacts and no customer result tied to them. The customer quotes it carries are about Claude generally. The page also does not say how the information barriers are enforced or audited, so a firm that must show a client an ethical wall will need to ask for that detail. Four of the artifacts were built by firms and vendors that sell into Claude's ecosystem.

What to do

• Run the cite check on a set of past filings where you already know the bad citations, and count what it misses before it touches live work.
• Ask Anthropic or your account contact how sharing settings and ethical walls are enforced, and what is logged, before connecting a client matter database.
• Borrow the pattern for non-legal review work: a first pass where every output links to its source and a named person signs off.