Orpaq AI

Zero-API-key lead generation platform — Discovers businesses, analyzes websites, and scores leads using local tools. No external API keys required.

Building this in the open — if it saves you time, a star helps others find it. Found a bug, or have an idea? Open an issue. New here? Start with a good first issue. New lead sources and scoring ideas welcome.

What it does

1. Discovers businesses from Google Maps (name, phone, website)
2. Analyzes websites for quality (PageSpeed, mobile-friendliness)
3. Scores leads (HOT/Medium/Low) based on upgrade opportunity
4. Outreach — generates personalized messages for each lead
5. Dashboard — Clean, performant, and minimal UI to view, filter, export leads, and trigger WhatsApp outreach directly via local bridge (port 10086).

Everything runs locally. No API keys, no subscriptions.

Quick start

# 1. Start the backend
cd backend
python3 -m uvicorn app.main:app --host 0.0.0.0 --port 8000

# 2. Start the frontend
cd frontend
npm run dev

# 3. Discover leads
python3 ~/.hermes/scripts/orpaq-research.py full "Kochi" "gyms" --max 20

Open http://localhost:3001 (frontend) or hit http://localhost:8000/api/v1 (API).

Tech Stack

• Backend: FastAPI + SQLite
• Frontend: Next.js 16 + Tailwind CSS
• Discovery: Playwright (Google Maps scraping)
• Analysis: Playwright performance API
• Scoring: Local algorithm (no AI API needed)

How it works

Google Maps (Playwright) → SQLite DB → FastAPI → Next.js Dashboard
         ↓
   Place page extraction (phone, website, rating)
         ↓
   Website analysis (PageSpeed, mobile check)
         ↓
   Lead scoring (HOT if no website)

Data

• Database: ~/.orpaq/leads.db (SQLite, auto-created)
• Suggestions auto-generated from your existing data
• No external API calls — everything runs on your machine

Pipeline

# Full pipeline: discover + analyze + score
python3 ~/.hermes/scripts/orpaq-research.py full "Kochi" "dentists" --max 20

# Discover only (just scrape Maps)
python3 ~/.hermes/scripts/orpaq-research.py discover "Kochi" "gyms" --max 10

# Check what needs enrichment
python3 ~/.hermes/scripts/orpaq-research.py needs-enrich

# Save a lead manually
python3 ~/.hermes/scripts/orpaq-write.py lead '{"business_id":"<id>","lead_score":95,"priority":"hot"}'

Environment

No API keys needed. The .env file is optional — defaults work out of the box.

Variable	Default	Purpose
SQLITE_PATH	~/.orpaq/leads.db	Database location