Search

More Filters

Show More
Add Listing
Go Back

CodeFuse muAgent

0
New
Multi-Agent Platforms & Frameworks
Report

Report Abuse

Basic Information

muAgent is a developer-focused agent framework that combines large language models with an Eventic Knowledge Graph (EKG) to enable multi-agent orchestration and SOP automation. It is designed to let teams build, debug, and run coordinated agent workflows using technologies like FunctionCall, CodeInterpreter, and RAG. The project offers both backend services (EKG services runnable via Docker Compose) and an SDK (pip install codefuse-muagent) to integrate LLMs and embedding models. The framework emphasizes human-in-the-loop control, online collaboration, and knowledge-on-demand by representing intentions, workflows, tools, and characters in a knowledge graph. Releases describe features such as one-click deployment, Java-based tool and agent registration, parallel EKG execution, and a canvas-based drag-and-drop front-end for designing agent scenarios. The README also points to demos, examples, and documentation for hands-on use and validation in complex DevOps scenarios at Ant Group.

Links

Stars
768
App URL
https://github.com/codefuse-ai/CodeFuse-muAgent
Github Repository
https://github.com/codefuse-ai/CodeFuse-muAgent/blob/main/README.md

Categorization

App Details

Features
EKG Builder for importing and modeling large volumes of documents and defining virtual teams, scene intentions, and semantic nodes for automated SOP workflows. EKG Assets include rich KG schema elements such as Intention Nodes, Workflow Nodes, Tool Nodes, and Character Nodes to improve tool selection, parameter filling, and human-agent collaboration. EKG Reasoning enables LLM-guided exploration under human supervision and captures successful exploration as knowledge entries. Diagnose provides visual editing, quick debugging, execution path recording, and end-to-end monitoring to reduce token usage and accelerate inference. Memory implements unified message pooling, retrieval, rerank and distillation to support long-context and multi-agent scenarios. ActionSpace follows Swagger for tool registration, categorization, permission management and offers a guarded code execution environment targeting safe function calling and code generation. The project also supplies an SDK, ekg-sdk integration, and parallel execution support.
Use Cases
muAgent helps teams design, run, and maintain complex multi-agent workflows by turning SOPs and documents into executable knowledge graph assets, enabling repeatable automation with human oversight. The framework lowers friction for deployment with Docker Compose recipes, one-click deployment features, and a drag-and-drop front-end for non-expert configuration. Developers can use the provided SDK and examples to connect LLMs and embedding models and register tools or agents, including Java-based integrations. Visual diagnostics and automatic documentation of successful execution paths reduce unnecessary LLM calls and token costs while improving reliability. Memory and message pooling improve context management for long-running or multi-agent interactions. The repository includes demos and was validated in real-world DevOps scenarios, illustrating use for operational automation, collaborative workflows, and rapid prototyping such as a text-based multiplayer demo.

Categories

Content Creation (5042)
Monetization & Business (2246)
Development Tools (1407)
Marketing & Social Media (1403)
Education & Career (1380)
Multi-Agent Platforms & Frameworks (1009)
Health & Life (957)
AI Resources (920)
Agents & Automation (778)
Data & Analytics (764)
Productivity Tools (679)
Writing & Editing (666)

Similar Listings

listing image

virtual lab

0 (0 review)
listing image

mnemo

0 (0 review)
listing image

EdgeChains

0 (0 review)
listing image

RAG Agents Accelerator

0 (0 review)
listing image

LLM Zero to Hundred

0 (0 review)

Featured Listings

Tags

Please fill the required fields*