AI for law firms that prepares the work while the lawyer decides

Case file summaries, contract review against the firm's templates, dates and deadlines, first drafts of letters: AI agents prepare this work in a space reserved for the firm. The lawyer reviews, edits and signs. Every run shows its cost.

  • An isolated space for the firm
  • The lawyer reviews and approves everything
  • A visible cost for every run
Aurore Studio
Skills/Service agreement review
Dropped fileclient-service-agreement-v3.docx
  1. Reading the agreement (24 pages)
  2. Comparing with the firm's template
  3. Flagging changed or missing clauses
  4. Extracting dates, terms and notice periods
Resultservice-agreement-review.docx
Duration 3 minCost €0.92Done

Where the firm's hours go

A large share of a lawyer's time goes into reading and formatting, before any actual analysis starts.

Reading the file

Going through dozens of documents to find out who said what, and when.

Comparing contracts

Reading an incoming contract line by line to see where it departs from the firm's template.

Tracking deadlines

Noting due dates, notice periods and key dates, matter after matter, without missing one.

What your agents prepare

Each workflow is built with you by the ERHA team, on your templates and real documents. It produces a working document, never an opinion.

Case file summaries

Reads a batch of documents (PDF, Word, scans), sorts them by date and type, and writes a summary in the firm's layout.

Contract review

Compares a contract with your templates and lists changed, added or missing clauses, with page references.

Dates and deadlines

Extracts dates, terms, notice periods and due dates from a document into a table, with the source of each date.

Letters from templates

A first draft of a letter in Word, from your templates and the facts of the matter, for the lawyer to rework.

Client emails

An agent drafts the email from your mailbox, you review it, and the client's reply comes back into the task.

Client follow-up

Client records with the history of emails and meetings, and a status summary of the relationship written on request.

How it gets set up

You build nothing yourself. You describe the work, the ERHA team builds it, and the lawyer keeps the final say.

  1. 01You describe the workflowIn a guided interview you explain the task and drop examples: a contract, your template, the memo you expect back.
  2. 02The ERHA team builds itThe workflow is developed and tested on your documents. You approve the specification before it goes live.
  3. 03You run it on a matterBy dropping the documents into the space or by asking an agent. The working document comes back with its cost, and the lawyer reviews it.

An example: a contract to review before signing

A five-lawyer firm receives a thirty-page service agreement from a client, drafted by the other side, to review within two days.

  1. The lawyer drops the agreement into the firm's space.
  2. The workflow compares it with the firm's template and lists the clauses that depart from it.
  3. Dates, terms and notice periods are extracted into a table, with their page.
  4. The lawyer reads the memo, writes her analysis and drafts the letter to the client from the template.

Outcome

The lawyer starts her review with the list of differences instead of a blank page.

Frequently asked questions

Does ERHA give legal advice?

No. A workflow produces a working document: a summary, a list of differences, a table of dates, a first draft. Analysis and advice stay with the lawyer, who reviews and approves everything.

Are the firm's documents kept separate?

Each space is isolated on every request: no other space can read your documents. Any sign-in by the ERHA team to your space is written to an audit log.

Can we choose which AI model is used?

Yes. The model is chosen per step or per agent, from OpenAI, Anthropic, Mistral AI, Google Gemini and others. Switching does not require rewriting the workflow.

How long does it take to get started?

A space takes a few minutes to create. A workflow is scoped in an interview, then the ERHA team builds it on your documents. You can also book a demo to be guided.

How is this different from ChatGPT?

With ChatGPT you paste your documents and instructions every time. An ERHA workflow is built once on your templates, runs on each matter with the same method, and shows its cost.

Hand over your first piece of work this week

Open your space in a few minutes, or book a demo to scope a first use case with the ERHA team.