Quantum Code Quick Start Guide

🚀 What is Quantum Code?

Quantum Code is an enterprise-grade AI orchestration platform that enhances code analysis through multi-model intelligence. Instead of relying on a single AI model, it orchestrates multiple AI providers (OpenAI GPT, Anthropic Claude, Google Gemini) to deliver more accurate, comprehensive code reviews and insights.

Key Innovation

Parallel AI Analysis: Runs multiple AI models simultaneously
Consensus Validation: Cross-verifies results across different models
Security-First: OWASP Top 10 vulnerability detection
Enterprise Ready: Production-grade architecture

📦 Installation

Option 1: From PyPI (Recommended)

pip install quantum-code

Option 2: From Source

git clone https://github.com/Codewithevilxd/quantum-code.git
cd quantum-code
pip install -e .

Verify Installation

# Check CLI
quantum --help

# Or using Python module
python -m quantum_code.cli --help

🔑 API Key Setup

Quantum Code requires API keys from at least one AI provider. Create a .env file in your project directory or home folder.

Step 1: Create Environment File

# In your project directory
cp .env.example .env

# Or in your home directory for global use
# ~/.quantum_code/.env

Step 2: Add API Keys

OpenAI (Recommended for beginners)

OPENAI_API_KEY=sk-proj-your-openai-key-here

Anthropic Claude (Best for code analysis)

ANTHROPIC_API_KEY=sk-ant-api03-your-anthropic-key-here

Google Gemini (Fast and cost-effective)

GEMINI_API_KEY=AIzaSy-your-gemini-key-here

Multiple Providers (Recommended)

# Primary providers
OPENAI_API_KEY=sk-proj-...
ANTHROPIC_API_KEY=sk-ant-api03-...
GEMINI_API_KEY=AIzaSy-...

# Optional: Enterprise providers
AZURE_OPENAI_API_KEY=...
AWS_ACCESS_KEY_ID=...
GOOGLE_CLOUD_PROJECT=your-project

Step 3: Verify Keys

# Test configuration
python -c "from quantum_code.settings import settings; print('Keys loaded successfully')"

🎯 Usage Modes

Mode 1: CLI Code Review (Standalone)

Basic Code Review

# Review a single file
quantum src/main.py

# Review entire directory
quantum src/

# Review specific file types
quantum src/ --include "*.py" "*.js"

Advanced Options

# Use specific AI model
quantum src/ --model claude-sonnet

# Use multiple models
quantum src/ --models gpt-4o,claude-sonnet,gemini-pro

# JSON output for CI/CD
quantum src/ --json > review-results.json

# Verbose logging
quantum src/ -v

# Specify project root
quantum src/ --base-path /path/to/project

Mode 2: MCP Server (Claude Code Integration)

Step 1: Configure Claude Code

Add to your ~/.claude.json:

{
  "mcpServers": {
    "quantum": {
      "command": "python",
      "args": ["-m", "quantum_code.server"]
    }
  }
}

Step 2: Restart Claude Code

# Restart Claude Code to load MCP server
claude

Step 3: Use in Claude Chat

Can you quantum codereview this authentication module?
Can you quantum compare these two architecture approaches?
Can you quantum debate the best testing strategy?

🛠️ Available Tools

1. Code Review (`codereview`)

Comprehensive code analysis covering:

Quality: Code structure, readability, maintainability
Security: OWASP Top 10 vulnerabilities, injection risks
Performance: Algorithm efficiency, resource usage
Architecture: Design patterns, scalability concerns

Usage:

# CLI
quantum src/auth.py

# Claude Code
"quantum codereview this authentication module for security issues"

2. Chat (`chat`)

Interactive AI assistance with:

Context Awareness: Repository and file understanding
Multi-turn Conversations: Maintains conversation history
Code Examples: Provides relevant code snippets

Usage:

# Claude Code
"quantum chat how does this authentication flow work?"
"quantum chat suggest improvements for this API design"

3. Compare (`compare`)

Side-by-side analysis of:

Architecture Approaches: Different design patterns
Technology Choices: Framework comparisons
Implementation Strategies: Various solutions

Usage:

# Claude Code
"quantum compare REST API vs GraphQL for this use case"
"quantum compare different state management approaches"

4. Debate (`debate`)

Advanced consensus building:

Multi-agent Debate: Models argue different perspectives
Step 1: Independent analysis from each model
Step 2: Cross-examination and critique
Final Consensus: Weighted recommendation

Usage:

# Claude Code
"quantum debate the best database choice for this application"
"quantum debate microservices vs monolith architecture"

⚙️ Configuration

Model Aliases

Use short names instead of full model identifiers:

Alias	Full Model	Provider
`mini`	gpt-4o-mini	OpenAI
`gpt`	gpt-4o	OpenAI
`sonnet`	claude-sonnet-4.5	Anthropic
`haiku`	claude-haiku-4.5	Anthropic
`gemini`	gemini-pro	Google
`flash`	gemini-flash	Google

Custom Configuration

Create ~/.quantum_code/config.yaml:

version: "1.0"
models:
  my-custom-model:
    litellm_model: openai/gpt-4o
    aliases:
      - custom
    notes: "My custom GPT-4o configuration"

🔍 Example Workflows

Scenario 1: New Feature Code Review

# 1. Write your code
# 2. Run comprehensive review
quantum src/new_feature.py --models claude-sonnet,gpt-4o

# 3. Address issues found
# 4. Run final verification
quantum src/new_feature.py

Scenario 2: Architecture Decision

# In Claude Code:
"quantum compare using Redis vs PostgreSQL for session storage"
"quantum debate the best caching strategy for this high-traffic API"

Scenario 3: Security Audit

# Security-focused review
quantum src/auth.py src/api.py --focus security

# OWASP compliance check
quantum src/ --security-scan

Scenario 4: CI/CD Integration

# .github/workflows/code-review.yml
name: AI Code Review
on: [pull_request]

jobs:
  review:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - name: Quantum Code Analysis
        run: |
          pip install quantum-code
          quantum src/ --json --models claude-sonnet,gpt-4o > review.json

🐛 Troubleshooting

"No API key found"

# Check if .env file exists
ls -la .env

# Verify key format
cat .env | grep OPENAI_API_KEY

# Test key loading
python -c "from quantum_code.settings import settings; print(settings.openai_api_key[:10] + '...')"

"Model not available"

# Check available models
quantum --list-models

# Update to supported model
quantum src/ --model gpt-4o-mini

MCP Server not connecting

# Test MCP server directly
python -m quantum_code.server

# Check Claude Code config
cat ~/.claude.json

# Restart Claude Code completely

Performance Issues

# Use fewer models for faster results
quantum src/ --models gpt-4o-mini,gemini-flash

# Use single model for quick checks
quantum src/ --model gemini-flash

📊 Understanding Results

Code Review Output

🔍 Code Quality: 8.5/10
✅ Good separation of concerns
✅ Clear naming conventions
⚠️  Consider adding input validation

🛡️ Security Score: 9.2/10
✅ No SQL injection vulnerabilities
✅ Proper authentication checks
⚠️  Add rate limiting for API endpoints

⚡ Performance: 7.8/10
✅ Efficient algorithms used
⚠️  Consider caching for frequent queries

Consensus Indicators

High Confidence: All models agree
Medium Confidence: Majority agreement with minor differences
Low Confidence: Significant disagreement (requires human review)

🎯 Best Practices

For Development Teams

Use multiple models for important reviews
Set up CI/CD integration for automated reviews
Configure model aliases for team consistency
Review consensus results carefully

For Individual Developers

Start with single model for quick feedback
Use compare mode for architecture decisions
Leverage chat mode for implementation guidance
Run security scans before deployment

Cost Optimization

Use appropriate models for task complexity
Batch reviews instead of reviewing single files
Configure rate limits to control API usage
Use cached results when possible

📈 Advanced Features

Custom Model Integration

Add proprietary or specialized models:

# ~/.quantum_code/config.yaml
version: "1.0"
models:
  my-local-llm:
    provider: cli
    cli_command: ollama
    cli_args: ["run", "codellama"]
    aliases: ["local"]

Enterprise Deployment

For organizations with custom requirements:

Private model endpoints
Custom security policies
Audit logging integration
SLA-based support

🤝 Support & Community

Getting Help

Documentation: https://github.com/Codewithevilxd/quantum-code
Issues: https://github.com/Codewithevilxd/quantum-code/issues
Discussions: GitHub Discussions tab

Enterprise Support

For commercial support and custom integrations:

Email: enterprise@codewithevilxd.com
Features: Custom model integration, enterprise deployment, training

Quantum Code - Multi-Model AI Orchestration for Superior Code Analysis

Ready to supercharge your code reviews? 🚀

pip install quantum-code
quantum --help

FilesExpand file tree

QUICKSTART.md

Latest commit

History

QUICKSTART.md

File metadata and controls

Quantum Code Quick Start Guide

🚀 What is Quantum Code?

Key Innovation

📦 Installation

Option 1: From PyPI (Recommended)

Option 2: From Source

Verify Installation

🔑 API Key Setup

Step 1: Create Environment File

Step 2: Add API Keys

OpenAI (Recommended for beginners)

Anthropic Claude (Best for code analysis)

Google Gemini (Fast and cost-effective)

Multiple Providers (Recommended)

Step 3: Verify Keys

🎯 Usage Modes

Mode 1: CLI Code Review (Standalone)

Basic Code Review

Advanced Options

Mode 2: MCP Server (Claude Code Integration)

Step 1: Configure Claude Code

Step 2: Restart Claude Code

Step 3: Use in Claude Chat

🛠️ Available Tools

1. Code Review (codereview)

2. Chat (chat)

3. Compare (compare)

4. Debate (debate)

⚙️ Configuration

Model Aliases

Custom Configuration

🔍 Example Workflows

Scenario 1: New Feature Code Review

Scenario 2: Architecture Decision

Scenario 3: Security Audit

Scenario 4: CI/CD Integration

🐛 Troubleshooting

"No API key found"

"Model not available"

MCP Server not connecting

Performance Issues

📊 Understanding Results

Code Review Output

Consensus Indicators

🎯 Best Practices

For Development Teams

For Individual Developers

Cost Optimization

📈 Advanced Features

Custom Model Integration

Enterprise Deployment

🤝 Support & Community

Getting Help

Enterprise Support

1. Code Review (`codereview`)

2. Chat (`chat`)

3. Compare (`compare`)

4. Debate (`debate`)