Agentic Lead Engine Ecosystem

AI-powered real estate lead generation with enforced commission collection, buyer-intent scoring, and automated outreach.

What It Does

A modular ecosystem of 7 single-purpose agents that source, score, contact, and convert real estate leads — while automatically tracking commissions, late fees, and agency subscriptions.

Why We Built It

Real estate agencies lose 40-60% of potential deals because leads are unqualified, follow-up is inconsistent, and commissions are tracked in spreadsheets. We built this to replace gut-feeling lead selection with data-driven scoring and to enforce commission collection programmatically.

Gaps It Fills

• No systematic lead scoring — agencies pick leads randomly; we score 0-100 with 6 measurable factors
• Manual outreach at scale — WhatsApp templates and automation replace hours of DMs
• Commission leakage — late fees, 60-day suspension, and TDS deduction enforced in code
• Fragmented tools — one framework covers sourcing, qualification, outreach, sales, analytics, conversion, and finance
• No referral routing — Model Y routing ensures receiving agency pays platform fee for qualified leads

The Ecosystem

Repo	Agent	What It Solves
Leads Agent	lead_agent_v2	Fetches leads from CSV, API, and real estate portals (Magicbricks, 99acres, Housing.com) with auto source selection
Qualification Agent	qualification_agent_v2	0-100 scoring across contact info, legitimacy, budget, timeline, digital presence, and source quality
Outreach Agent	outreach_agent_v2	Personalized WhatsApp outreach templates and message composer for agencies
Sales Agent	sales_agent_v2 + followup_agent_v2	Handles objections, parses replies, pushes CTAs, and schedules automated follow-ups
Analytics Agent	analytics_agent_v2 + optimization_agent_v2	Funnel metrics, conversion rates, and AI-driven optimization suggestions
Conversion Agent	conversion_agent_v2	Dynamic pricing engine, deal routing, and referral Model Y activation
Commission Agent	commission_agent.py	10%/3% splits, Net 15 invoicing, 2%/week late fees, 60-day suspension, 1% TDS

What More Can Be Done

• Replace mock/scraped leads with live Meta Lead Ads and Google Local Services Ads integrations
• Add machine learning for conversion likelihood prediction based on historical closure data
• Launch React Native mobile CRM with swipe-based lead management
• Integrate Razorpay for automated invoice generation and payment reconciliation
• Expand to 5+ Indian cities with localized lead sources and pricing tiers
• Build agency dashboard with real-time funnel visualization and agent performance leaderboards

Quick Start

git clone https://github.com/MrHTC/AGENTIC-LEAD-ENGINE.git
cd AGENTIC-LEAD-ENGINE
cp .env.example .env
pip install -r requirements.txt
python run.py cycle real_estate Delhi

Demo

Architecture

AGENTIC-LEAD-ENGINE/
├── agents/
│   ├── lead_source_adapter.py       # CSV / API / mock / real-estate portal sources
│   ├── qualification_agent_v2.py    # 6-factor scoring + tiers
│   ├── outreach_agent_v2.py         # WhatsApp templates
│   ├── commission_agent.py          # Splits, late fees, suspension
│   ├── analytics_agent_v2.py        # Metrics
│   ├── conversion_agent_v2.py       # Pricing + referral routing
│   ├── followup_agent_v2.py         # Follow-up scheduling
│   ├── sales_agent_v2.py            # Reply handling
│   ├── lead_agent_v2.py             # Validation
│   └── optimization_agent_v2.py     # Suggestions
├── orchestrator/
│   └── orchestrator.py              # Main pipeline
├── api/
│   ├── mobile_crm.py                # CRM logic
│   └── mobile_crm_server.py         # Flask server
├── utils/
│   ├── csv_memory.py                # Memory tables: leads, contacted, replies, conversions, commissions, invoices
│   ├── logger.py                    # Logging
│   ├── ollama_client.py             # AI classification
│   └── whatsapp_sender.py           # WhatsApp integration
├── config/
│   └── settings.py                  # Environment settings
├── docs/
│   ├── contracts/                   # Service agreement, lead buyer agreement
│   ├── legal/                       # Terms of service, privacy policy
│   └── policies/                    # Commission, late payment, quality SLA
├── requirements.txt
├── .env.example
├── .gitignore
├── LICENSE
├── CODE_OF_CONDUCT.md
├── CONTRIBUTING.md
└── run.py                           # CLI entry point

Lead Scoring Model

Every lead gets a 0-100 score across:

• Contact info (0-20): phone, email, location
• Business legitimacy (0-20): GMB status, category, team size
• Financial indicators (0-15): price range, budget signals
• Engagement readiness (0-15): existing score, tags, timeline
• Market demand (0-15): location presence, niche match
• Digital presence (0-15): email, website

Tiers:

Tier	Score range
COLD	0-59
WARM	60-84
HOT	85-100

Commission & Payment Rules

• On deal closure: platform gets 10%, source partner gets 3%
• Invoice due Net 15
• Late fee: 2% per week after 15-day grace
• Suspension: after 60 days overdue
• TDS: 1% under Section 194H (India)

Mobile CRM API

Run the API server:

python run.py api

Endpoints (all require X-API-KEY or ?api_key= if MOBILE_CRM_API_KEY is set):

• GET /api/health
• GET /api/leads
• GET /api/leads/<id>
• PATCH /api/leads/<id>/status
• POST /api/leads/<id>/notes
• POST /api/leads/<id>/followups
• GET /api/leads/<id>/analytics
• GET /api/dashboard

Requirements

• Python 3.9+
• Flask (for API mode)
• See requirements.txt

Community

• Code of Conduct
• Contributing
• License

Topics

real-estate lead-generation automation python whatsapp sales analytics conversion qualification commission crm buyer-intent ai flask api