Hexabot vs mcp-agent
AI workflow and agent frameworks compared
Hexabot and mcp-agent help developers build AI agents and workflows that use tools and conversational capabilities. Hexabot combines YAML-defined automation with chat channels in a TypeScript runtime, while mcp-agent is a Python framework centered on MCP integrations and composable agent workflows.
Hexabot: Build AI Workflows for Conversational Channels
Hexabot is a TypeScript platform for building and running agentic workflows, actions, and conversational channels in one runtime. It suits teams that want to connect AI tools and bots through configurable workflows rather than build a runtime from scratch.
mcp-agent: Build Agents with Model Context Protocol
A Python framework for building LLM agents around Model Context Protocol servers and composable workflow patterns. It suits developers who need MCP integrations, multi-agent workflows, or durable execution with Temporal.
| Hexabot | mcp-agent | |
|---|---|---|
| Language | TypeScript | Python |
| License | — | Apache-2.0 |
| Stars | 1.1k | 8.6k |
| Forks | 240 | 893 |
| Last analyzed | Oct 10, 2026 | Oct 3, 2026 |
Key differences
- Hexabot uses TypeScript and YAML-defined workflows; mcp-agent uses Python and provides composable workflow patterns such as routers and orchestrator-workers.
- Hexabot combines workflow automation with conversational channels, actions, memory, and a data layer; mcp-agent focuses on agents that use capabilities exposed through MCP servers.
- mcp-agent offers asyncio execution and optional Temporal durability for pause, resume, and retries; Hexabot describes runtime memory integration and database options including SQLite and Postgres.
- Hexabot's CLI creates and manages projects, with generated projects specifying Node.js ^24.17.0; mcp-agent's CLI scaffolds and deploys projects, and its examples use asynchronous patterns.
- Hexabot's license metadata is NOASSERTION, while its README states FCL-1.0-ALv2; mcp-agent lists Apache-2.0.
- Both projects are unarchived; the supplied figures list 1.3k stars and 246 forks for Hexabot, and 8.6k stars and 893 forks for mcp-agent.
Choose Hexabot if you…
- need YAML-defined automations that connect actions, tools, and memory with conversational channels.
- want a TypeScript runtime with a CLI and SQLite or Postgres data-layer options.
- plan to prototype a locally runnable workflow and can review the license terms and runtime setup.
Choose mcp-agent if you…
- need a Python framework for connecting agents to MCP servers.
- want to compose agent patterns or use Temporal for durable workflows with pause, resume, and retries.
- need to expose an application or agent workflow as an MCP server.
Compare the numbers side by side, or add more projects
This comparison is generated with AI from the OSRepos analyses of both projects. Always check each project's repository and documentation before choosing.