Level the playing field. Rent-Swarm is an advanced AI dashboard that empowers tenants with institutional-grade intelligence to find, analyze, and negotiate their next lease. Access LIVE at https://api.rent-swarm.tech
Rent-Swarm is composed of specialized AI agents working in concert:
- 🕵️ The Scout: An autonomous scraper that finds listings (Craigslist integrated), extracts hidden data (like image IDs), and assigns a "Scam Score" and "True Cost" estimate.
- ⚖️ The Lawyer: A RAG-powered legal analyst. Upload a PDF lease, and it flags risky clauses, illegal terms, and red flags based on local jurisdiction laws.
- 📈 The Forecaster: Predictive analytics engine. Tells you if a unit is overpriced based on historical data and market trends.
- 💬 The Negotiator: Generates professional, leverage-based negotiation emails. It uses data from the Scout and Forecaster to draft persuasive copy.
- 📂 Intelligence Dossier: Summarizes all findings for a specific listing into a downloadable, classified-style PDF report.
- Frontend: Next.js 14 (App Router), React, Tailwind CSS, Shadcn UI
- Backend: Next.js API Routes, Server Actions
- Database: MongoDB (User Data, Bookmarks)
- Authentication: NextAuth.js (Email/Password)
- AI Logic: Google Gemini 2.0 Flash (Reasoning), LangChain
- Scraping: Puppeteer, Browserbase (Headless Browser Infrastructure)
- PDF Generation: jsPDF (Client-side generation)
- Node.js 18+
- MongoDB Database (Atlas or Local)
- API Keys for Google Gemini, Browserbase
-
Clone the repository
git clone https://github.com/yourusername/rent-swarm.git cd rent-swarm/rent-swarm-dashboard -
Install dependencies
npm install
-
Environment Setup Create a
.env.localfile in the root directory:# Database MONGODB_URI=your_mongodb_connection_string # Auth NEXTAUTH_URL=http://localhost:3000 NEXTAUTH_SECRET=your_generated_secret # AI & Services GOOGLE_API_KEY=your_gemini_api_key BROWSERBASE_API_KEY=your_browserbase_key BROWSERBASE_PROJECT_ID=your_project_id
-
Run the Development Server
npm run dev
Open http://localhost:3000 to view the dashboard.
This project is licensed under the MIT License.