Open Source Intercom Alternatives

Chatwoot centralizes customer conversations into one inbox no matter where they arrive, covering website live chat, email, Facebook, Instagram, Twitter, WhatsApp, Telegram, Line and SMS. Support teams stop switching between a dozen consoles, and because the platform is self-hosted, the customer data stays under the company's own control.

Captain, the built-in AI agent, automates responses and handles common queries so routine questions resolve without an agent reading them, leaving the team on conversations that need judgment. A help center portal publishes articles, FAQs and guides, which trims the repetitive questions that reach the inbox in the first place.

The daily work of a support team is covered by a wide feature set.

• Collaboration: private notes, mentions, labels, canned responses, keyboard shortcuts and a command bar.
• Routing: auto-assignment by agent availability, teams, business hours, auto-responders and capacity management.
• Customer data: contact profiles with interaction history, segments, custom attributes, pre-chat forms and outbound campaigns.
• Integrations: Slack, Dialogflow, Shopify, Linear, Google Translate and dashboard apps embedded in the interface.
• Reporting: a live view of ongoing conversations plus conversation, agent, inbox, label, team and CSAT reports, downloadable for offline analysis.

Deployment includes a one-click Heroku button and a DigitalOcean Kubernetes app, with further options in the documentation, though environment variables need setting carefully or features break. Development follows a git flow model on the develop branch, so anyone wanting stability should track master or a tagged release.

Chaskiq puts the front line of customer service into software you run yourself: a web messenger on your site, agents answering it, campaigns going out and a help center sitting behind both. The project describes itself as source available rather than fully open, a distinction worth understanding before building on it.

A Rails backend serves a GraphQL API with OAuth authorization, consumed by a React frontend. PostgreSQL holds the records and Redis handles caching and transient data, so the operational shape is familiar to anyone who has run a Rails application before.

The platform reaches well past a chat widget.

• Web messenger: an embeddable chat with a rich text editor, animated GIFs, embedded video, syntax-highlighted code and video calls over WebRTC.
• Segments and routing: customers are filtered by custom attributes and conversations routed to the right agent.
• Automation: triggerable conversational bots, onboarding tours and mailing campaigns run without an agent present.
• Help center: a multilingual article system that lives on a custom domain or inside the messenger itself.
• Integrations: WhatsApp, Twitter DM, Slack, Calendly, Zoom, Pipedrive and webhooks connect it to the rest of your stack.
• Governance: configurable GDPR consent, audits on access and record changes, and a composable roles and permissions system.

Development guides exist for macOS, Ubuntu, Windows and Docker, needing Ruby 2.6 or newer, PostgreSQL 10 or newer and Redis 2.6 or newer. A commercial license lifts the AGPL and commons clause terms for proprietary use.

Cossistant ships a support chat widget as React components you compose, rather than a script tag that injects someone else's interface into your page. The approach is code-first and API-driven, with headless hooks and primitives, real-time messaging, and the backend to run all of it, documented for developers and the AI tools they use.

Two packages carry the client side. The React SDK provides the headless hooks and primitives, and a companion package adds Next.js bindings and utilities. Behind them sits a Turborepo monorepo running on Bun, with Hono and tRPC for the API, Drizzle ORM for the database, Better Auth for sessions, WebSockets for live messages, and Docker running Postgres and Redis locally.

The split between widget and infrastructure shapes what you actually get.

• Headless components: styling and layout stay yours, because the SDK ships behavior rather than a fixed look.
• Real-time messaging: conversations travel over WebSockets instead of polling an endpoint.
• Full backend: the API, database and auth are part of the project, not a hosted dependency you rent.
• Next.js package: a separate module covers the framework-specific bindings and utilities.
• Documentation: a quickstart for adding it to an app and a contributors guide.

The project is licensed AGPL-3.0 for non-commercial use, and commercial deployments need a license from the maintainers. It is part of the Vercel open source program and runs a program of its own for other open source projects. It suits a team that wants support chat in their product without handing the conversation to a third party.

Written in Node.js and Express, Tiledesk is a live chat platform with chatbots built in, for teams running their own customer support instead of renting a widget. Conversations can be automated with its native chatbot technology, connected to other systems through APIs and webhooks, and given entire visual applications to run inside the chat itself.

Its multichannel model removes the duplicated work of maintaining one bot per channel. Chatbot scripts are written once with images, buttons and whatever elements a channel supports, and replies auto-adapt to wherever they land. This repository is the dashboard, an Angular back office, while the chat components come from the Chat21 project.

The back office covers the day-to-day of a support desk.

• Conversations: unlimited chats with labels, notes, canned responses and full history.
• Routing: departments and agent groups direct conversations, with operating hours respected.
• Ticketing: a ticketing system sits beside live chat for anything that outlives a session.
• Widget: the web widget is customizable, with iOS and Android apps available as separate components.
• Reporting: data export, analytics and a built-in CRM track what the desk is handling.

The quickest install is the published Docker Compose setup, which brings up every component at once. From source you need Node, the Angular CLI and a running Tiledesk server, with settings held in environment files or a remote dashboard config. The dashboard is MIT licensed, and the cloud service offers the same modules under the same license.