Skip to main content
Kaino.dev
Discover
Evals
News
Academics
Insights
Kaino.dev

Discover, evaluate, and compare AI tools, models, and agents.

Explore

  • Discover
  • Evaluations
  • News
  • Academics
  • Insights

Community

  • Twitter
  • YouTube
  • Instagram
Privacy PolicyTerms of Service

© 2026 Kaino.dev. All rights reserved.

Version 1.1.0
MiroFlow · Discover · Kaino
Discover/AI AGENTS/MiroFlow
MiroFlow logo

AI AGENTS

MiroFlow

by MiroMindAI

agentlead-sourcegithub-agent-framework-repositoriessource:github.comagentic-aiai-agentstool-integrationmcpbenchmarksweb-uigithub
Visit WebsiteDocumentationGitHub

Overview

MiroFlow is an agentic AI platform for building intelligent agents with tool integration, benchmark evaluations, and a web UI.

Details

MiroFlow is described by its documentation as an agentic AI platform for building intelligent agents with tool integration, benchmark evaluations, and quick-start documentation. Its core concepts cover workflows for intent recognition, planning, sub-agent delegation, MCP tool access, and result synthesis. The GitHub repository excerpt also highlights a web UI, benchmark performance, and support for MiroThinker, Claude, Kimi, and OpenAI.

When to Use

Build or evaluate agentic AI workflows that need planning tool access sub-agent delegation and result synthesis. Experiment with an agent platform that documents benchmark evaluations and provides quick-start material. Explore a web-UI-based agent framework with support for MiroThinker Claude Kimi and OpenAI.

Getting Started

  1. Open the official documentation at https://miromindai.github.io/MiroFlow/ and review the quick-start material.
  2. Read the core concepts page to understand intent recognition
  3. planning
  4. sub-agent delegation
  5. MCP tool access
  6. and result synthesis.
  7. Inspect the GitHub repository at https://github.com/MiroMindAI/MiroFlow for project files
  8. setup instructions
  9. and examples before production use.

Key Features

  • •Agentic AI platform for building intelligent agents
  • •Tool integration
  • •including MCP tool access described in the core concepts documentation
  • •Workflow concepts for intent recognition
  • •planning
  • •sub-agent delegation
  • •and result synthesis
  • •Web UI highlighted in the GitHub repository excerpt
  • •Benchmark evaluation focus highlighted in the documentation and repository excerpt
  • •Support for MiroThinker
  • •Claude
  • •Kimi
  • •and OpenAI noted in the repository excerpt

Capabilities

  • •agentic-workflows
  • •tool-integration
  • •planning
  • •sub-agent-delegation
  • •mcp-tool-access
  • •benchmark-evaluation
  • •web-ui

Last updated Jun 4, 2026