hexabot-ai/Hexabot

Hexabot v3 is an AI workflow automation platform, combining workflows, actions, agents, and conversational channels in one runtime.

What it solves

Hexabot is an automation platform designed to handle "boring" repetitive tasks by combining AI agents, structured workflows, and multi-channel communication into a single runtime. It allows developers to build agentic workflows that can interact with various tools and channels while maintaining state and context.

How it works

The platform uses a schema-first architecture based on Zod for validation. It allows users to define workflows in YAML, which are then executed as typed runtime contracts. The system separates task logic from configuration through a binding system, integrates the Model Context Protocol (MCP) for tool interoperability, and provides explicit memory definitions to maintain context across interactions.

Who it’s for

It is built for developers who want to create AI-driven automation and agentic workflows that can be deployed across different communication channels.

Highlights

  • Agentic Workflows: Defined via YAML with typed runtime contracts.
  • MCP Integration: Support for the Model Context Protocol for better tool and context interoperability.
  • Multi-channel Support: Maintains continuity across different conversational channels.
  • Memory and RAG: Built-in support for memory definitions and retrieval-augmented generation.
  • Action-based Execution: Uses schema-validated inputs and outputs to define behavior.