codefuse-chatbot  by codefuse-ai

AI assistant for software development lifecycle, powered by multi-agent framework

created 1 year ago
1,210 stars

Top 33.1% on sourcepulse

GitHubView on GitHub
Project Summary

CodeFuse-ChatBot is an open-source AI assistant designed to streamline the entire software development lifecycle, from design to operations. It targets developers and DevOps professionals by integrating a multi-agent framework with DevOps toolkits, code and document RAG, and a sandbox environment, enabling LLMs to handle complex tasks efficiently.

How It Works

The system leverages a Multi-Agent Schedule Core for intelligent task orchestration, supporting various configurations. It incorporates Multi-Source Web Crawling for data acquisition, a Data Processor for document handling, and Text Embedding & Indexing for efficient retrieval. Data is managed via Vector and Graph Databases, with Prompt Control and Management defining agent context. A secure Sandbox environment allows code execution, all orchestrated by a flexible LLM core supporting open-source models and APIs.

Quick Start & Requirements

  • Install: pip install codefuse-muagent
  • Prerequisites: NVIDIA drivers, Python 3.9.18, CUDA 11.7. For Apple Silicon, brew install qpdf may be needed.
  • Setup: Configure server_config.py, then run bash start.sh in the examples directory or use start.py.
  • Docs: CodeFuse-muAgent, Quick Start Details

Highlighted Details

  • Warehouse-level code understanding and project file-level code generation.
  • Document analysis enhanced by knowledge bases and knowledge graphs.
  • Customizable knowledge base construction for the DevOps domain.
  • Compatibility with smaller, domain-specific models for DevOps platforms.

Maintenance & Community

The project is initiated by the Ant Group CodeFuse team. Contributions are welcomed via GitHub Issues. Further details can be found in the Contribution Guide.

Licensing & Compatibility

The project is open-source, with specific licensing details not explicitly stated in the README for commercial use or closed-source linking.

Limitations & Caveats

The README mentions testing on Windows and x86 macOS, with potential considerations for Apple Silicon. Specific LLM integrations like codellama-34b-int4 are listed as "coming soon."

Health Check
Last commit

1 year ago

Responsiveness

1 week

Pull Requests (30d)
0
Issues (30d)
1
Star History
45 stars in the last 90 days

Explore Similar Projects

Feedback? Help us improve.