whatisgithub

What is fancy-core?

kuohuafan/fancy-core — explained in plain English

Analysis updated 2026-05-18

11HTMLAudience · developerComplexity · 4/5LicenseSetup · hard

In one sentence

A self deployable case management and AI workbench for law firms, with an open core plus separately licensed billing and tax modules for lawyers only.

Mindmap

mindmap
  root((Fancy AI))
    What it does
      Case management
      Role based permissions
      AI drafting workbench
    Tech stack
      Next.js
      Prisma
      PostgreSQL
    Use cases
      Self host firm case system
      Choose your own AI provider
      Extend with custom plugins
    Audience
      Law firms
      Lawyers and paralegals
    Licensing
      Open core CPAL
      Restricted lawfirm module
      Paid legal data service

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

What do people build with it?

USE CASE 1

Self host a case management and AI drafting system for a law firm.

USE CASE 2

Assign six built in roles across thirteen permissions to match firm structure.

USE CASE 3

Choose between Claude, Gemini, OpenAI, or Mistral for AI drafting using your own keys.

USE CASE 4

Extend the workbench with custom plugins through the Fancy JavaScript API.

What is it built with?

Next.jsPrismaPostgreSQLDockerNode.js

How does it compare?

kuohuafan/fancy-core1000sbin/moa-goods-makerbrunosimon/esin-e1-p2026
Stars111111
LanguageHTMLHTMLHTML
Last pushed2022-05-15
MaintenanceDormant
Setup difficultyhardeasyeasy
Complexity4/51/51/5
Audiencedevelopergeneralgeneral

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · hard Time to first run · 1h+

Requires Node.js 20+, PostgreSQL, Prisma migrations, and at least one AI provider API key.

So what is it?

Fancy AI is a self deployable management system built for law firms, combining case management, permission control, day to day administrative workflow, and AI features into one workbench. A law firm can run the system and its database on infrastructure it controls, and choose which AI provider handles the AI features: Anthropic Claude, Google Gemini, OpenAI, or Mistral AI, using the firm's own account and API key for each. The README is explicit that in this self hosted setup, the project's maintainers do not collect model usage fees and never hold the firm's provider keys. The repository draws a clear line between what is open and what is paid. The core tracked in this repository, including case handling, permissions, the AI workbench, a plugin API, and a few example tools like a deadline calculator and a conflict of interest checker, is released under the CPAL-1.0 open source license, which allows use, modification, and deployment but requires sharing source code changes and keeping a required branding notice. Separate from this core, a law firm specific module covering billable hours, invoicing, receipts, performance tracking, document scanning, bookkeeping, and tax filing is offered under a different, more restrictive license: free for lawyers and law firms to use and modify for themselves, but they cannot resell it or offer it as a service to others. Legal reference data and retrieval features, along with deployment, integration, and training services, are separate paid offerings not included in this repository. The system supports six built in roles, such as firm partner, administrator, lawyer, paralegal, accountant, and consultant, across thirteen distinct permissions like viewing case files, drafting documents, running the calendar, and handling billing, so access can be tuned to how a real firm is organized. Branding can be customized with the firm's own name and logo, applied across login, sidebar, and generated documents, while the required attribution notice must still remain visible. Technically, the full self hosted version runs on Node.js 20 or later with PostgreSQL and Prisma for the database layer, and can be started with Docker Compose or directly with npm. A separate single file demo version runs entirely in the browser using temporary session storage for testing the interface, but the README warns it should never be used with real client data or long lived API keys. The README repeatedly stresses that all AI generated drafts, summaries, or legal citations are only aids for research and must be reviewed by a qualified professional before use.

Copy-paste prompts

Prompt 1
Help me set up fancy-core locally with Docker Compose and Prisma migrations.
Prompt 2
Explain how fancy-core's role based permission system controls the thirteen listed permissions.
Prompt 3
Show me how to write a custom plugin using Fancy.registerPage and Fancy.ai.
Prompt 4
Walk me through configuring the .env file to connect fancy-core to Claude and Gemini.

Frequently asked questions

What is fancy-core?

A self deployable case management and AI workbench for law firms, with an open core plus separately licensed billing and tax modules for lawyers only.

What language is fancy-core written in?

Mainly HTML. The stack also includes Next.js, Prisma, PostgreSQL.

How hard is fancy-core to set up?

Setup difficulty is rated hard, with roughly 1h+ to a first successful run.

Who is fancy-core for?

Mainly developer.

Open on GitHub → Ask about another repo

This repo across BitVibe Labs

Verify against the repo before relying on details.