Eko AI Business Automation

Autonomous AI Agents for Prospecting, Outreach, Sales & Deal Closure
Developed by Eko AI Automation LLC (Denver, CO)

English

Overview

Eko AI is a multi-agent sales automation platform built for local businesses. It discovers leads, enriches them with AI-powered research, reaches out via email and voice, and manages the entire pipeline from first contact to closed deal — all with built-in compliance (TCPA, CAN-SPAM, CPA Colorado, FCC AI Rule).

Architecture

A 6-agent LangGraph system orchestrates the full sales cycle:

Agent	Function
Discovery	Geo-targeted prospecting (Google Maps, LinkedIn, Yelp, Colorado SOS)
Research	Lead enrichment with NLP, gap analysis, and dynamic scoring
Outreach	Multi-channel contact: email, voice AI, SMS
CRM	Pipeline management, dynamic lead scoring, adaptive follow-ups
Customer Success	Onboarding, health monitoring, churn prevention
Compliance	TCPA consent validation, CAN-SPAM unsubscribe, CPA Colorado, FCC AI Rule

Architecture Diagram

flowchart TB
    subgraph Sources[Discovery Sources]
        GM[Google Maps]
        YP[Yelp]
        LI[LinkedIn]
        SOS[Colorado SOS]
    end

    subgraph Agents[AI Agents]
        DA[DiscoveryAgent]
        RA[ResearchAgent]
        OA[OutreachAgent]
        CA[ComplianceAgent]
    end

    subgraph Channels[Channels]
        EM[Email / Resend]
        VA[Voice / VAPI.ai]
        TG[Telegram]
    end

    subgraph Storage[Storage]
        PG[(PostgreSQL + pgvector)]
        RD[(Redis)]
        ST[/Static Files /]
    end

    subgraph Frontend[Frontend]
        NX[Next.js 14 + Tailwind]
        US[User Dashboard]
        LP[Landing Page]
    end

    User -->|Login| NX
    NX -->|API Calls| FA[FastAPI Backend]
    FA -->|Celery Tasks| CW[Celery Worker]
    FA -->|Cache/Queue| RD
    FA -->|Persist| PG
    FA -->|Serve| ST

    Sources -->|Raw Leads| DA
    DA -->|Enrich| RA
    RA -->|Score & Analyze| PG
    FA -->|Trigger| OA
    OA -->|Send| EM
    OA -->|Call| VA
    OA -->|Notify| TG
    CA -->|Validate| OA

    CW -->|Scheduled| DA
    CW -->|Follow-ups| OA
    CW -->|DNC Sync| PG

    Paperclip[Paperclip AI] -->|Governance| FA

    classDef source fill:#e1f5fe,stroke:#01579b
    classDef agent fill:#fff3e0,stroke:#e65100
    classDef channel fill:#e8f5e9,stroke:#2e7d32
    classDef storage fill:#f3e5f5,stroke:#6a1b9a
    classDef frontend fill:#fce4ec,stroke:#c2185b
    class GM,YP,LI,SOS source
    class DA,RA,OA,CA agent
    class EM,VA,TG channel
    class PG,RD,ST storage
    class NX,US,LP frontend

Tech Stack

Backend: Python 3.11, FastAPI, LangGraph, SQLAlchemy (async), Celery
Frontend: Next.js 14 (App Router), Tailwind CSS, TanStack Query
Database: PostgreSQL 15 + pgvector
Cache / Queue: Redis
AI / LLM: OpenAI GPT-4o, MiniMax M2.7, Kimi Code
Email: Resend (outbound + inbound webhooks)
Voice: VAPI.ai (outbound + inbound assistant "Eva")
Scraping: Outscraper, Apify, SerpApi
Calendar: Custom booking page with Google Calendar integration
Observability: Paperclip AI (agent governance & audit)

Quick Start

Requirements

Docker & Docker Compose
Node.js 20+ (for local frontend development)
API keys: OpenAI, Resend, Outscraper (see .env.example)

1. Clone & configure

git clone https://github.com/enderjnets/Eko-AI-Business-Automation.git
cd Eko-AI-Business-Automation
cp .env.example .env
# Edit .env with your API keys

2. Start services

docker compose up -d

This starts:

PostgreSQL on localhost:5432
Redis on localhost:6379
Backend API on http://localhost:8000
Frontend on http://localhost:3001

3. Verify

curl http://localhost:8000/health

4. Access the dashboard

Open http://localhost:3001 in your browser.

API Endpoints

Leads

GET /api/v1/leads — List leads
POST /api/v1/leads — Create lead manually
GET /api/v1/leads/{id} — Get lead
PATCH /api/v1/leads/{id} — Update lead
POST /api/v1/leads/{id}/enrich — Enrich with ResearchAgent
POST /api/v1/leads/discover — Run DiscoveryAgent

Campaigns

GET /api/v1/campaigns — List campaigns
POST /api/v1/campaigns — Create campaign
POST /api/v1/campaigns/{id}/launch — Launch campaign
POST /api/v1/campaigns/{id}/pause — Pause campaign

Emails

POST /api/v1/emails/{lead_id}/send — Send email
POST /api/v1/emails/{lead_id}/generate-and-send — AI-generate and send

Voice

POST /api/v1/voice-agent/calls — Place outbound AI call

Analytics

GET /api/v1/analytics/pipeline — Pipeline summary
GET /api/v1/analytics/performance — Performance metrics

Project Structure

├── backend/
│   ├── app/
│   │   ├── agents/           # LangGraph agents (Discovery, Research, Outreach, CRM, Compliance)
│   │   ├── api/v1/           # FastAPI routers
│   │   ├── db/               # SQLAlchemy models & migrations
│   │   ├── models/           # Database models
│   │   ├── schemas/          # Pydantic schemas
│   │   ├── services/         # External API integrations (Resend, VAPI, etc.)
│   │   └── tasks/            # Celery background tasks
│   └── tests/
├── frontend/
│   ├── app/                  # Next.js App Router
│   ├── components/           # React components
│   └── lib/                  # API client & utilities
├── docker-compose.yml
└── docs/

Compliance

TCPA: Consent logging, pre-contact validation, DNC registry sync
CAN-SPAM: Functional unsubscribe in every email, opt-out honoring
CPA Colorado: Right to access, deletion, and opt-out
FCC AI Rule: AI disclosure in calls and emails

License

Español

Resumen

Eko AI es una plataforma de automatización de ventas multi-agente construida para negocios locales. Descubre leads, los enriquece con investigación impulsada por IA, contacta por email y voz, y gestiona todo el pipeline desde el primer contacto hasta el cierre — todo con cumplimiento normativo integrado (TCPA, CAN-SPAM, CPA Colorado, FCC AI Rule).

Arquitectura

Sistema multi-agente construido con LangGraph que orquesta todo el ciclo de ventas:

Agente	Función
Discovery	Prospección geolocalizada (Google Maps, LinkedIn, Yelp, SOS CO)
Research	Enriquecimiento de leads con NLP y análisis de brechas
Outreach	Contacto multicanal: email, voz, SMS
CRM	Pipeline, lead scoring dinámico, seguimiento adaptativo
Customer Success	Onboarding, monitoreo de salud, prevención de churn
Compliance	Validación TCPA, CAN-SPAM, CPA Colorado, FCC AI Rule

Stack Tecnológico

Backend: Python 3.11, FastAPI, LangGraph, SQLAlchemy, Celery
Frontend: Next.js 14, Tailwind CSS, TanStack Query
Database: PostgreSQL 15 + pgvector
Cache/Queue: Redis
LLM: OpenAI GPT-4o, MiniMax M2.7, Kimi Code
Email: Resend
Voz: VAPI.ai (asistente inbound "Eva")
Scraping: Outscraper, Apify, SerpApi
Calendario: Página de booking propia con integración Google Calendar
Gobernanza: Paperclip AI

Inicio Rápido

Requisitos

Docker & Docker Compose
Node.js 20+ (para desarrollo frontend local)
API keys: OpenAI, Resend, Outscraper (ver .env.example)

1. Clonar y configurar

git clone https://github.com/enderjnets/Eko-AI-Business-Automation.git
cd Eko-AI-Business-Automation
cp .env.example .env
# Editar .env con tus API keys

2. Levantar servicios

docker compose up -d

Esto inicia:

PostgreSQL en localhost:5432
Redis en localhost:6379
Backend API en http://localhost:8000
Frontend en http://localhost:3001

3. Verificar

curl http://localhost:8000/health

4. Acceder al dashboard

Abre http://localhost:3001 en tu navegador.

Fases de Desarrollo (MVP Rápido)

Fase	Semana	Entregable
1	1-2	Discovery + Research + Dashboard
2	3-4	Email Outreach + CRM Pipeline
3	5-6	Voice AI + Calendar integration
4	7-8	Customer Success + Analytics
5	9-10	Compliance + Production deploy

Cumplimiento Normativo

TCPA: Registro de consentimientos, validación previa a contacto, sincronización DNC
CAN-SPAM: Unsubscribe funcional en todos los emails
CPA Colorado: Derecho de acceso, borrado y opt-out
FCC AI Rule: Divulgación de IA en llamadas y emails

Licencia

Contact / Contacto: eko@ekoai.com | Denver, CO

Name		Name	Last commit message	Last commit date
Latest commit History 267 Commits
.github/workflows		.github/workflows
.kimi		.kimi
backend		backend
docs		docs
frontend		frontend
scripts		scripts
.env.example		.env.example
.gitignore		.gitignore
AGENTS.md		AGENTS.md
CHANGELOG.md		CHANGELOG.md
KIMI_SWARM_PROMPT.md		KIMI_SWARM_PROMPT.md
LICENSE		LICENSE
Makefile		Makefile
README.md		README.md
SECURITY.md		SECURITY.md
docker-compose.main.yml		docker-compose.main.yml
docker-compose.yml		docker-compose.yml

Folders and files

Latest commit

History

Repository files navigation