A command-line interface tool that uses AI to assist with various tasks including code analysis, command generation, and more.
This codebase was completely developed by Claude Code (Anthropic's Claude AI assistant). The author is not responsible for any issues that may arise from its use.
We are planning to continue developing this codebase using an agent-only approach, with minimal human intervention.
- Command generation and execution with safety validation
- Code analysis and insights
- Content collection and summarization
- Directory structure analysis
- Rich terminal output formatting
# Clone the repository
git clone https://github.com/your-username/aiagent.git
cd aiagent
# Build the application
go build -o aiagent cmd/aiagent/main.go# Basic usage
./aiagent "your request here"
# Use mock LLM (no API key needed)
./aiagent --mock "your request here"
# Enable verbose mode
./aiagent -v "your request here"
# Force approve commands (use with caution)
./aiagent -y "your request here"# Generate and execute a command
./aiagent "show me the disk usage in a readable format"
# Analyze code
./aiagent "collect all information about the formatter component"
# Get a direct answer
./aiagent "explain what chmod does"By default, the application uses the OpenAI API. You need to set the OPENAI_API_KEY environment variable:
export OPENAI_API_KEY="your-api-key"This project is open source and available under the MIT License.