kevinrivm/vocero-crm

CRM de WhatsApp open source y self-hosted con agente de IA y Laboratorio de auto-evaluacion

What it solves

Vocero CRM provides a self-hosted, privacy-focused customer relationship management system specifically for businesses using WhatsApp. It solves the problem of managing high-volume WhatsApp communication while integrating AI agents that are rigorously tested for quality before interacting with real customers.

How it works

The platform acts as a central hub for WhatsApp conversations, featuring a real-time inbox, a sales pipeline, and an integrated AI agent. Its standout feature is the "Laboratory," a sandbox environment where simulated customer personas (e.g., angry customers, price-sensitive buyers) interact with your AI agent. An independent LLM judge evaluates these conversations, providing a performance score, identifying hallucinations or knowledge gaps, and suggesting improvements. For advanced users, the CRM allows you to disable the built-in agent and connect your own via a dedicated API, ensuring your WhatsApp tokens remain secure within your own infrastructure.

Who it’s for

  • AI/Automation Agencies: Professionals who deploy CRM solutions for clients and need a way to prove the quality of their AI agents using the Laboratory's reporting features.
  • Businesses: Companies that want to automate WhatsApp customer service using AI without relying on third-party SaaS providers that store their sensitive customer data.

Highlights

  • Laboratory Sandbox: Automated testing of AI agents against diverse customer personas with actionable feedback and scoring.
  • Real-time WhatsApp Inbox: Supports rich media, real-time updates, and seamless handoff to human agents.
  • Self-Hosted Privacy: All data, attachments, and AI agent knowledge bases remain on your own server.
  • Flexible Integration: Use the built-in AI agent or connect your own custom microservice via a secure API.
  • Pipeline & Analytics: Built-in Kanban board for lead management and conversion tracking, including optional integration with Meta's Conversions API.

Related

  • Project
    eddyzzl/marvis-risk-agent

    A local-first, governed credit-risk Agent platform that transforms natural language requests into audited risk analysis, models, and strategies.

  • Project
    DEEP-JLU/Awesome-Graph-Engineering

    A curated collection of research papers, benchmarks, and open-source projects on graph engineering for LLM agents, organized into model, individual, and system intelligence levels. It accompanies an arXiv survey and serves as a living index for the community.

  • Project
    moss-site/moss-trade-bot-skills

    Moss is an open‑source framework that turns natural‑language descriptions of crypto trading styles into fully‑featured quantitative bots. It includes a local backtesting engine, an automated weekly evolution loop that refines tactical parameters, safety guardrails, and optional integration with the Moss platform for verification and live trading. Install via a tagged release, let the AI infer indicators and risk settings, run evolution backtests, and (if desired) bind the bot to the Moss service for real‑time execution.

  • Project
    openvort/openvort

    An open-source platform for deploying autonomous AI agents that integrate with enterprise messaging apps to perform tasks like code review, CI/CD management, and reporting.