Back to directory

Hexabot

An AI workflow automation platform that combines YAML-defined workflows, actions, agents, and conversational channels, with tools, MCP, memory, and retrieval-augmented generation, in one runtime.

Hexabot

Hexabot v3 is an automation platform with first-class AI capabilities, combining workflows, actions, and conversational channels in one runtime. Workflows are defined in YAML with typed runtime contracts, actions define workflow behavior with schema-validated inputs, outputs, and settings, and a binding system separates reusable capability and configuration bindings from task logic; the platform supports explicit memory definitions, MCP integration for tool and context interoperability, and multi-channel continuity across conversational surfaces.

A project is created and run through a CLI installed with npm, which scaffolds a new project, auto-detects the package manager, and starts a development server with an admin UI, an API, and non-production API documentation each on their own local endpoint; hexabot create requires an interactive terminal for its initial admin-credential prompts. The CLI also handles environment configuration, health checks, and database migrations.

Its schema-first architecture uses Zod broadly for validation and shared contracts across the platform.

More in Agent frameworks

See all 902
  • Recurse

    recurse.run

    A serverless harness that lets a coding agent design, test, and deploy its own specialist sub-agents as tools, MCP servers, or bots, defined by a schema-validated manifest.

  • ProofForge

    shulgin.is-a.dev

    An AI-agent pipeline that decomposes a mathematics problem, proves each piece, and formalizes the result in Lean 4 and Mathlib until the Lean kernel accepts it, with six of its proofs already merged into Google DeepMind's formal-conjectures repository.

  • local-coder

    github.com

    Builds a local coding-agent team on top of Ollama and OpenCode by detecting the machine's hardware, recommending models for seven specialist roles, and testing whether each assigned model can actually use tools, edit a repository and pass verification.