Wenlan
Wenlan은 소스 기반 AI 지식베이스·LLM 위키로, 문서·결정을 페이지화해 검색·검토·리뷰 가능
저장소 정보는 아직 수집하지 않았습니다.
아직 감지된 업데이트가 없습니다. GitHub 릴리스가 나오면 여기에 보입니다.
Instant AI analysis for contract's summary, risks & verdict in seconds.
README 전체를 불러오는 중입니다.
README 전문 보기 ↗Legalyze AI is an AI-powered legal document analyzer. Upload a contract, NDA, or agreement, and a multi-agent AI pipeline summarizes it, flags risky clauses, and returns a safe/unsafe verdict — streamed to the browser in real time.
This is a two-service app: a Next.js frontend/app-server and a separately-deployed FastAPI analysis backend.
backend/README.md├── src/ # Next.js app (see CLAUDE.md for the full directory map)
│ ├── app/ # routes, incl. app/api/* — the only code allowed to talk to the DB or backend directly
│ ├── components/ # UI, split by page/domain
│ └── lib/ # auth (better-auth), db (Drizzle), services, utils
├── drizzle/ # DB schema migrations
├── backend/ # FastAPI analysis service — deployed independently, see backend/README.md
└── public/ # static assets
backend/README.md) or pointed at a deployed instancepnpm install
Create a .env in the project root:
# Database
DATABASE_URL="postgresql://..."
# better-auth
BETTER_AUTH_URL="http://localhost:3000"
BETTER_AUTH_SECRET="..."
GOOGLE_CLIENT_ID="..."
GOOGLE_CLIENT_SECRET="..."
GITHUB_CLIENT_ID="..."
GITHUB_CLIENT_SECRET="..."
# FastAPI backend
WEB_URL="http://127.0.0.1:8000" # or your deployed FastAPI Cloud URL
# Shared secret sent to the backend. Server-side only, never NEXT_PUBLIC_.
# Must equal BACKEND_API_SECRET on the backend. Generate: openssl rand -hex 32
BACKEND_API_SECRET="..."
# Bug report emails (src/app/api/report/route.ts)
SMTP_HOST="..."
SMTP_PORT="587"
SMTP_USER="..."
SMTP_PASS="..."
SMTP_FROM="..."
TO_EMAIL="..."
# Optional — Google Analytics
NEXT_PUBLIC_GA_ID=""
pnpm test # frontend (Vitest)
cd backend && uv run pytest # backend (pytest, runs offline)
npx drizzle-kit generate # generate a migration from schema changes
npx drizzle-kit migrate # apply migrations
pnpm dev
Runs on http://localhost:3000. Make sure the backend (backend/) is running too — the frontend calls it directly for all analysis requests.
legalyze-ai.vercel.appbackend/README.md for deployment detailsContributions are welcome! Please open an issue or submit a pull request.
This project is licensed under the MIT License.