AI Agents Orchestrator

•
101
•

Coordinates diverse AI coding assistants to collaborate on complex software development tasks through sequential workflows or a multi-agent team runtime.

AI Agents Orchestrator provides two independent systems designed to enhance software development: an Orchestrator for step-based workflows (e.g., implement, review, refine) and an Agentic Team runtime that enables role-based AI agents (like Project Managers, Developers, and QAs) to communicate freely and collaboratively solve problems. It integrates various cloud and local AI coding assistants such as Claude, Codex, Gemini, Copilot, Ollama, and llama.cpp. The platform boasts an enterprise-grade agentic infrastructure, including 9 specialized agents, a library of 22 reusable skills, over 34 Model Context Protocol (MCP) tools for various functions, and a sophisticated graph-based context memory system utilizing hybrid search for persistent learning across sessions.

주요 기능

01Provides two independent systems: a step-based Orchestrator and a free-communication Agentic Team runtime.
02Features a graph-based context system with hybrid search (BM25 + semantic) for persistent memory and learning.
03Includes a comprehensive agentic infrastructure with 9 specialized agents, 22 reusable skills, and 34+ MCP tools.
04Orchestrates multiple AI coding assistants (Claude, Codex, Gemini, Copilot, Ollama, llama.cpp).
0528 GitHub stars
06Offers both a REPL/CLI interface and a Vue/Nuxt UI dashboard for interaction and visualization.

사용 사례

01Automating multi-stage software development processes from implementation and coding to review and refinement.
02Facilitating collaborative problem-solving among a team of specialized AI agents for complex programming projects.
03Building and managing enterprise-grade AI agent systems that learn and adapt across multiple development sessions.