Hexabot – Open-source AI Chatbot & Agent Platform

TL;DR

TL;DR: Hexabot is an open-source chatbot and AI agent platform that combines visual workflow building with MCP tool integration, multi-channel support, and RAG — set up in minutes via the CLI.

Source and Accuracy Notes

⚠️ This section is MANDATORY. All links must be verified from actual source, not guessed.

• Project page: hexabot.ai ← MUST visit and verify
• Source repository: github.com/Hexastack/Hexabot ← MUST read README
• License: FCL-1.0-ALv2 (proprietary, see LICENSE.md)
• HN launch thread: news.ycombinator.com/item?id=41724497
• Source last checked: 2026-07-18 (commit main)

What Is Hexabot?

Hexabot is an open-source automation platform with first-class AI capabilities. It combines visual chatbot building, AI agent workflows, multi-channel support (web, REST API, GraphQL), and extensions via MCP (Model Context Protocol) tools, memory, and RAG (Retrieval-Augmented Generation).

The project describes itself as:

Build and run agentic workflows across channels with YAML, tools, MCP, memory and RAG.

Key capabilities include:

• Visual flow builder — define workflows via YAML or the CLI
• MCP integration — connect to external tools and services through the Model Context Protocol
• Multi-channel — web widgets, REST API, GraphQL endpoints
• AI memory — persistent context across conversations
• RAG support — connect to vector databases for document-grounded responses
• Extensions system — documented at hexabot.ai/extensions
• Docker support — run everything in containers with docker compose

The current version is 3.3.4 (pnpm monorepo).

Setup Workflow

Prerequisites

• Node.js ^24.17.0
• One package manager: npm, pnpm, yarn, or bun
• Docker (optional, for Docker-based services)

Step 1: Install the CLI

npm install -g @hexabot-ai/cli

Or run without a global install:

npx @hexabot-ai/cli --help

Step 2: Create a project

hexabot create my-project
cd my-project
hexabot dev

The hexabot create command auto-detects your package manager. To force one explicitly:

hexabot create my-project --pm npm

During creation, the CLI prompts for initial admin credentials and requires an interactive terminal (TTY).

Default local endpoints are shown after hexabot dev starts.

Step 3: Run with Docker (optional)

From the project root:

docker compose up

Deeper Analysis

Hexabot sits in the same space as tools like Botpress, Flowise, and Voiceflow — but with first-class MCP support and a CLI-first workflow that suits developer environments.

The v3 architecture is a pnpm monorepo with separate packages under packages/. The MCP integration is the most differentiated feature compared to older chatbot builders — it lets Hexabot agents call external tools at runtime without custom webhook code.

The FCL-1.0-ALv2 license is a proprietary license. The core platform is open-source and runnable self-hosted, but commercial use may require a separate agreement with Hexastack.

Practical Evaluation Checklist

• [ ] CLI installs with npm install -g @hexabot-ai/cli
• [ ] hexabot create scaffolds a working project
• [ ] hexabot dev starts the local server without errors
• [ ] Docker Compose starts all services
• [ ] MCP tools are discoverable in the dashboard
• [ ] Web widget renders on a test page
• [ ] RAG pipeline connects to a vector DB (pgvector, Qdrant, etc.)

Security Notes

• Admin credentials are set at project creation — use strong passwords in production
• The Docker Compose setup exposes ports locally; do not expose to the internet without a reverse proxy (nginx, Cloudflare Tunnel)
• MCP tool permissions should be scoped — avoid running Hexabot with overly broad system tool access
• Check SECURITY.md for the current security policy and reporting process

FAQ

Q: What LLM providers does Hexabot support? A: Hexabot integrates with any LLM that speaks OpenAI-compatible APIs. MCP tools can also bridge to providers like Anthropic, Google Gemini, and self-hosted models.

Q: Can I self-host Hexabot? A: Yes. The monorepo includes a Docker Compose configuration for self-hosting without relying on Hexastack’s cloud.

Q: Is this free to use commercially? A: The source is open under FCL-1.0-ALv2. Review the LICENSE.md carefully — commercial deployments may require a separate agreement with Hexastack.

Q: Does Hexabot require a database? A: Yes, it uses PostgreSQL for persistence (configured via Docker Compose out of the box).

Q: How does MCP integration work? A: MCP (Model Context Protocol) lets Hexabot agents call external tools at runtime. You configure MCP servers in the project settings, and tools become available as actions in the visual flow builder.

Conclusion

Hexabot is a capable open-source chatbot and AI agent platform with strong developer ergonomics. The CLI-first setup gets you from zero to a running bot in under five minutes, and MCP integration means you can connect to external tools without building custom webhooks. If you need a self-hosted alternative to commercial chatbot platforms, it is worth a look — just review the FCL-1.0-ALv2 license before commercial deployment.