TrendWhat rose on GitHub, every day at 07:30 KST · 한국어

GitHub · as of November 8, 2026 · View on GitHub →

melgarafael/DeskcommCRM

Self-hosted AI sales CRM

+250 stars in 7 days

Run a self-hosted AI sales operating system for WhatsApp businesses

Imagine an AI agent that reads customer messages, qualifies them, and moves the sales pipeline on its own. This tool lets you run that AI agent and CRM on a single server. It requires no monthly subscription, keeps your data on your own server, and serves as an alternative to commercial services like Kommo or Intercom.

How the AI handles sales

The core of this system is that the AI is not just a chatbot but an agent that manipulates the CRM. The AI converses with customers to gather information, moves leads through pipeline stages, and hands off to humans when necessary. It uses organization-specific RAG (Retrieval-Augmented Generation) to leverage knowledge relevant to the business, and all AI actions are recorded in an audit log for traceability. The system also offers a feature where the AI learns and suggests performance improvements, a process that always requires human approval.

Simplicity of installation and operation

Installation is completed with a single command if Docker is installed on your VPS. The installation script automatically sets up all necessary components, including issuing HTTPS certificates and applying the database schema. Updates can be performed with a single button click from the interface, and a safety mechanism automatically rolls back to the previous version if an update fails. For users unfamiliar with server management, a guide integrated with an AI coding assistant is provided, allowing installation to proceed simply by giving instructions in natural language.

Support for various business models

This CRM started in e-commerce but is currently used in various industries such as clinics, real estate, and information products. Pipeline stage names and terminology can be freely changed to fit the industry, allowing you to rename 'lead' to 'patient' or 'buyer', and 'success' to 'payment completed' or 'booking confirmed'. WhatsApp integration supports two methods: QR code scanning and the official Meta API, and includes built-in spam prevention features. Additionally, webhooks allow external tools to push leads directly into the pipeline, and conditional automation rules can be set to optimize workflow.

Data sovereignty and security

All data is stored on the user's VPS, and a multi-tenant structure based on Supabase ensures data isolation between organizations. To comply with LGPD (Brazilian General Data Protection Law), data anonymization and audit log retention policies are applied by default. Server-side RBAC (Role-Based Access Control) allows for fine-grained management of permissions for each team member, and two-factor authentication is also supported. This tool is open source, and all features are available for free without locking any specific functionality.

Things to check before use

Before adopting this tool, a basic understanding of server management is required. You must install Docker on your VPS and prepare API keys for a Supabase account and an AI service (such as OpenAI or Anthropic). A VPS with at least 4GB of RAM is recommended, and the process of connecting a WhatsApp number via QR code is necessary. Also, since the behavior of the AI agent varies depending on the configuration, it is important to clearly define the AI's scope of authority and budget limits during initial setup. Since this tool assumes self-hosting, it may not be suitable for users who prefer cloud-based SaaS.

Written by AI from this repository's README on October 27, 2026. For the facts, the original on GitHub is the reference.

By the numbers

GitHub trending history
13 days between Jun 19, when Trend began recording, and Nov 8, 2026 · first on Sep 11 · best daily rank #2 (Sep 12)

Read from GitHub, as of November 8, 2026.

Language
TypeScript
Topics
ai · ai-agents · chatbot · crm · customer-support · kommo-alternative · lgpd · mcp
Stars
5,278
Stars gained in 7 days
+250
Contributors
218
Forks
1,507
Latest release
v1.104.0 · November 7, 2026
Commits
16,911
Last commit
November 8, 2026
Open issues
343
Open pull requests
163

Related repositories

View on GitHub → · Homepage

README badge

trend.ondo.sh · GitHub daily trending #2 · September 12, 2026

The badge above shows the highest this repository has ranked on GitHub's daily trending since June 19, 2026, and the day it did. Paste the Markdown below into your README. The badge opens this page, and it updates by itself when the repository climbs higher.

[![trend.ondo.sh](https://trend.ondo.sh/github/r/melgarafael--DeskcommCRM/badge.en.svg)](https://trend.ondo.sh/github/r/melgarafael--DeskcommCRM/en.html)