Skip to content

a low cost, high-workflow-efficiency, infinitely customizable platform for users to build their personal AI systems on top of in the terminal.

License

Notifications You must be signed in to change notification settings

Caronex-Labs/IntelligenceInterface

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Intelligence Interface

A self-evolving meta-system that serves as an intelligent orchestrator of AI capabilities, where Caronex acts as the central nervous system coordinating agents, spaces, and evolutionary processes.

Overview

Intelligence Interface is a revolutionary meta-system implemented as a Go-based TUI application that provides an intelligent interface to AI capabilities. The system features a unique architecture designed for evolution and self-improvement:

  • Caronex Manager Agent: Central intelligence providing coordination, planning, and delegation
  • Agent-Everything Architecture: Every capability implemented as specialized intelligent agents
  • Space-Based Computing Foundation: Prepared for persistent desktop environments that evolve through conversation
  • Comprehensive Tool System: Management tools for system introspection, coordination, and configuration
  • Bootstrap-Ready Architecture: Foundation for future system self-improvement capabilities

Sprint 1 Achievements

Sprint 1 has established a solid foundation for the Intelligence Interface meta-system:

✅ Core Infrastructure

  • Directory Migration: Organized codebase into logical agent-based architecture
  • Git Repository: Full version control with comprehensive change tracking
  • Configuration Foundation: Extended configuration system supporting meta-system requirements
  • BDD Infrastructure: Comprehensive testing framework with Godog integration

✅ Caronex Manager Agent

  • Central Orchestrator: Intelligent coordination of all system components
  • Management Tools: 5 specialized tools for system introspection and coordination
  • Agent Registry: Dynamic agent discovery and capability management
  • TUI Integration: Seamless mode switching with Ctrl+M hotkey

✅ Performance & Quality

  • Outstanding Performance: 25,666+ operations/30sec with 0% error rate
  • 100% BDD Compliance: All scenarios implemented and validated
  • 100% Technical Debt Resolution: Clean, maintainable codebase
  • Comprehensive Testing: Integration, performance, and stability test suites

Prerequisites

  • Go 1.24+ installed
  • API keys for AI providers (optional, but recommended)

Quick Start

1. Clone and Navigate

git clone [repository-url]
cd IntelligenceInterface

2. Install Dependencies

go mod download

3. Build the Application

go build -o ii

4. Run the Application

# Basic run - launches TUI interface
./ii

# Or run directly with go
go run main.go

Using Intelligence Interface

Agent Modes

Intelligence Interface features a dual-agent system:

🤖 Implementation Agent Mode (Default)

  • Purpose: Direct code implementation, file editing, analysis
  • When to use: Writing code, editing files, technical implementation tasks
  • Visual: Standard interface with implementation-focused tools

⚡ Caronex Manager Mode

  • Purpose: System coordination, planning, task delegation
  • When to use: Project planning, task coordination, system oversight
  • Access: Press Ctrl+M to switch to Caronex mode
  • Visual: Distinct visual styling with coordination-focused interface
  • Capabilities:
    • System introspection and status monitoring
    • Agent coordination and capability assessment
    • Task planning and delegation
    • Configuration inspection and validation
    • Space foundation management

Running Options

Debug Mode

# Run with debug logging
go run main.go -d

Specific Working Directory

# Run with specific working directory
go run main.go -c /path/to/your/project

Non-Interactive Mode

# Run in non-interactive mode with a prompt
go run main.go -p "your prompt here"

Configuration

API Keys Setup

For AI features to work, set up API keys as environment variables:

# Add to your ~/.bashrc, ~/.zshrc, or set before running
export ANTHROPIC_API_KEY="your-anthropic-key"
export OPENAI_API_KEY="your-openai-key"
export GOOGLE_API_KEY="your-google-key"
export GROQ_API_KEY="your-groq-key"

The system automatically selects appropriate models based on available API keys.

Configuration Files

The application uses cascading configuration:

  1. Global: ~/.ii.json
  2. Project: ./.ii.json
  3. Environment variables (highest priority)

Features

Terminal User Interface (TUI)

  • Interactive terminal interface using Bubble Tea framework
  • Multiple themes available
  • Intuitive navigation and command system

Multi-Provider AI Support

  • 9+ AI providers supported (OpenAI, Anthropic, Google, etc.)
  • Automatic provider selection based on available API keys
  • Streaming responses for better user experience

Agent System

  • Coder Agent: High-capability code generation and implementation
  • Summarizer Agent: Efficient content summarization
  • Title Agent: Creative title generation
  • Task Agent: Planning and task breakdown
  • Caronex Manager: System coordination, planning, and agent orchestration (✅ Implemented)

Session Management

  • Hierarchical sessions with parent-child relationships
  • Automatic summarization when approaching context limits
  • Persistent conversation history
  • Cost tracking across providers

Tool System

  • File operations (view, edit, write)
  • Shell execution (bash)
  • Code search (grep, glob)
  • LSP integration for code intelligence
  • Extensible tool framework

Testing

Intelligence Interface features a comprehensive testing framework with multiple test types:

Unit Tests

# Run all unit tests
go test ./...

# Run with verbose output
go test -v ./...

# Run specific package tests
go test -v ./internal/llm/prompt

BDD Tests (Godog Framework)

# Run BDD scenarios
go test ./test/bdd/... -v

# Run specific feature tests
go test ./test/bdd/... -v -godog.format=pretty

Integration Tests

# Run integration test suite
go test ./test/integration/... -v

# Skip long-running tests
go test ./test/integration/... -v -short

Performance Tests

# Run performance benchmarks
go test ./test/performance/... -v -bench=.

# Run performance baselines
go test ./test/performance/... -v

Test Coverage

# Generate coverage report
go test -cover ./...

# Generate detailed coverage
go test -coverprofile=coverage.out ./...
go tool cover -html=coverage.out

Development

Building for Production

go build -ldflags="-s -w" -o ii

Database Operations

# Generate database code from SQL (requires sqlc)
sqlc generate

# Database migrations are automatically applied on startup
# Migration files are in: internal/db/migrations/

Code Generation

# Go template generation (future feature)
go run templates/projects/go_backend_gorm/cmd/standardize/main.go --config user_domain.yaml

Architecture

Space-Based Meta-System Architecture

  • Caronex Core: Central orchestrator coordinating all system components
  • User Interface Space: TUI/CLI/API agents providing adaptive user interaction
  • Agent Management Space: Dynamic agent spawning, coordination, and evolution
  • Data Management Space: Session, memory, and cache agents with persistence
  • Evolution Space: Bootstrap compiler and golden repository integration

Key Components

  • Bubble Tea TUI: Responsive terminal interface
  • SQLite Database: Lightweight, embedded storage with automatic migrations
  • Model Context Protocol (MCP): Extensible AI tool connectivity
  • Language Server Protocol (LSP): Code intelligence integration

Troubleshooting

Build Failures

  • Ensure Go 1.24+ is installed: go version
  • Run go mod tidy to clean up dependencies

Missing Configuration

  • The application creates configuration files automatically
  • Check ~/.ii.json for global settings

API Errors

  • Verify API keys are set correctly in environment
  • Check provider-specific error messages in debug mode

Database Issues

  • SQLite database is created automatically at first run
  • Database location: ./ii.db (configurable)

Performance Issues

  • Use debug mode (-d) to identify bottlenecks
  • Check token usage in session management
  • Monitor provider response times

Architecture

Sprint 1 Implementation

Intelligence Interface follows a modular, agent-based architecture designed for evolution:

IntelligenceInterface/
├── internal/
│   ├── agents/              # Agent Implementation Layer
│   │   ├── base/           # Core agent framework
│   │   ├── builtin/        # Built-in agents (coder, summarizer, title, task)
│   │   └── caronex/        # Caronex manager agent ✅
│   ├── core/               # Core Infrastructure
│   │   ├── config/         # Extended configuration system ✅
│   │   └── logging/        # Structured logging
│   ├── tools/              # Tool Ecosystem
│   │   ├── builtin/        # Standard tools (file, shell, search)
│   │   └── coordination/   # Management tools for Caronex ✅
│   ├── services/           # Application Services
│   ├── infrastructure/     # Data & Communication
│   └── tui/               # Terminal User Interface ✅
├── test/                   # Comprehensive Test Framework
│   ├── bdd/               # BDD scenarios with Godog ✅
│   ├── integration/       # End-to-end integration tests ✅
│   └── performance/       # Performance benchmarks ✅
├── templates/             # Bootstrap Compiler Foundation
└── .claude/               # Meta-System Memory Bank ✅

Key Architectural Achievements

🏗️ Agent-Everything Foundation

  • All capabilities implemented as intelligent agents
  • Caronex Manager Agent provides central coordination
  • Agent registry with dynamic capability discovery
  • Clear separation between manager and implementation agents

🛠️ Management Tool System

  • System Introspection: Complete system state assessment
  • Agent Coordination: Task planning, delegation, and status monitoring
  • Configuration Inspection: Comprehensive configuration analysis
  • Agent Lifecycle: Agent readiness and capability management
  • Space Foundation: Future space management preparation

🔄 Space-Based Computing Readiness

  • Configuration system supports space definitions
  • Architecture prepared for persistent desktop environments
  • Agent-to-space mapping capabilities established
  • UI layout configuration framework in place

📊 Performance Excellence

  • Sub-millisecond response times for core operations
  • Zero-error stability under sustained load (25,666+ ops)
  • Concurrent access support with excellent performance
  • Memory-efficient design with comprehensive leak testing

The Vision

Intelligence Interface isn't just another productivity tool - it's an AI-powered desktop environment that evolves with you. Think of Spaces like different desktops on your Mac - each one evolved over time for the work you regularly do:

  • Knowledge Base Space: Like your personal library that has grown to understand exactly how you research and write
  • Development Space: Like your workshop that has accumulated all the tools and configurations you've perfected over years of coding
  • Social Space: Like your office setup that knows your team, your communication patterns, and your collaboration preferences

You switch between spaces with simple hotkeys, and each space remembers everything about how you like to work.

Contributing

See .claude/ directory for comprehensive development documentation including:

  • Project architecture and patterns
  • BDD workflows and testing
  • Technical debt tracking
  • Memory coordination system

License

[License information to be added]

Acknowledgments

Originally forked from Intelligence Interface - an open-source AI coding assistant.

About

a low cost, high-workflow-efficiency, infinitely customizable platform for users to build their personal AI systems on top of in the terminal.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Go 51.2%
  • Python 28.1%
  • Jinja 18.3%
  • Gherkin 1.0%
  • Just 0.8%
  • Makefile 0.3%
  • Other 0.3%