vexis-cli 1.0.7

AI command-line agent for terminal automation and task execution

VEXIS-CLI-1

AI-Powered Command Line Interface for Intelligent Automation

VEXIS-CLI-1 is an AI agent derived from VEXIS-1.1 that performs tasks through command execution. It leverages large language models to intelligently interpret natural language commands and execute them through terminal operations, enabling automated workflow management and system administration.

Quick Start • Documentation • Models • Configuration • Contributing

Key Features

AI-Powered Intelligence

• Natural Language Processing: Execute commands using plain English descriptions
• Context-Aware Execution: Understands your workflow and adapts to your needs
• Multi-Model Support: Compatible with 80+ AI models from 12 major providers
• Smart Verification: Automatic task completion validation with confidence scoring

Advanced Automation

• Two-Phase Engine: Planning and execution phases for reliable task completion
• Cross-Platform Compatibility: Works seamlessly on macOS, Linux, and Windows
• GUI Automation: Integrate terminal commands with graphical interface interactions
• Screenshot Integration: Visual context capture for enhanced understanding

Developer Experience

• Rich Terminal UI: Beautiful, informative output with progress indicators
• Flexible Configuration: YAML-based settings with environment variable overrides
• Extensible Architecture: Plugin-ready design for custom integrations
• Comprehensive Logging: Structured logging for debugging and monitoring

🚀 Quick Start

Prerequisites

• Python 3.8 or higher
• Ollama installed and running (for local AI models)
• Git (for cloning the repository)

Installation

1. Clone the repository

   git clone https://github.com/AInohogosya-team/VEXIS-CLI-1.git
   cd VEXIS-CLI-1
   

2. Install dependencies

   pip install -r requirements.txt
   

3. Set up Ollama (optional, for local models)

   # Install Ollama
   curl -fsSL https://ollama.com/install.sh | sh
   
   # Pull a recommended model
   ollama pull llama3.2:latest
   

4. Run VEXIS-CLI

   python run.py
   

Your First Command

# Start the interactive interface
python run.py

# Or use direct commands
vexis-cli "List all Python files in the current directory"

Supported AI Models

VEXIS-CLI-1 supports 150+ models from 20 major providers through Ollama:

Core Providers

• Meta: Llama 3.1/3.2 series (8B, 70B, 1B, 3B variants)
• Google: Gemma 2/3 series (1B-27B parameters, multimodal capabilities)
• DeepSeek: R1/V3/Coder series (8B-671B, reasoning and coding specialists)
• Microsoft: Phi-3/4 series (3.8B-14B, efficient small models)
• Mistral: Mistral/Large/Ministral series (7B-675B, European open-source leader)

Advanced Providers

• Alibaba (Qwen): Qwen 2.5/3 series (0.5B-235B, multilingual with 128K+ context)
• IBM: Granite/Code series (350M-34B, enterprise-grade models)
• BigCode: StarCoder 2 series (3B-15B, specialized for code generation)
• Cohere: Command R series (7B-35B, retrieval-augmented generation)
• 01.AI: Yi series (1.5B-34B, bilingual models)

Specialized Models

• Vision-Language: LLaVA, Moondream, Qwen3-VL (7B-235B)
• Coding: DeepSeek Coder, Qwen Coder, Granite Code, StarCoder 2
• Agentic: Hermes 3, Reflection, Devstral Small 2 (3B-405B)
• Cloud-Only: GPT-OSS, Gemini 3, GLM-5, MiniMax, Kimi (20B-744B)

Cloud & Local Models

• Local Models: Run entirely on your machine with Ollama
• Cloud Models: Access high-performance models via API
• Hybrid Mode: Seamlessly switch between local and cloud models

Complete Model List

Popular Local Models:

• llama3.2:latest (3B) - Balanced performance with 128K context
• qwen2.5:7b - Multilingual capabilities with 128K context
• mistral:7b - Fast and efficient with 32K context
• deepseek-r1:8b - Advanced reasoning with 128K context
• gemma2:9b - High-performing with 8K context
• phi3:mini - Efficient small model with 4K context

High-Performance Cloud Models:

• deepseek-v3:671b - State-of-the-art MoE with 160K context
• qwen3:235b - Advanced MoE with 256K context
• mistral-large-3:675b-cloud - Multimodal enterprise model
• gpt-oss:120b-cloud - Frontier performance
• gemini-3-flash-preview:cloud - Built for speed

Usage Examples

Quick Start

# Start the interactive interface
python run.py

# Or use direct commands
vexis-cli "List all Python files in the current directory"

For detailed usage examples, see our Detailed Guide.

Documentation

• 📖 Detailed Guide - Comprehensive usage examples and advanced features
• 🔧 Troubleshooting - Common issues and solutions
• 📚 API Reference
• 🏗️ Architecture
• ⚙️ Configuration
• 🤝 Contributing

Community

• GitHub Issues: Report bugs and request features
• Discussions: Join the community discussion

License

This project is licensed under the MIT License - see the LICENSE file for details.

Support

• 📧 Email: [e-mail]
• X: AInohogosya
• Home Page: https://ainohogosya.github.io/home-page/

---

Back to top

Made with ❤️ by the AInohogosya-team