Help & Documentation

🔍Step 1: Connect Your GitHub

Connect your GitHub account to enable code health analysis. Echeo will scan your repositories to analyze code quality, complexity, technical debt, and predict maintenance costs.

• Go to your Dashboard and click "Connect GitHub"
• Authorize Echeo to access your repositories
• We'll automatically scan your codebase (this takes a few minutes)
• Your code health metrics will appear as analysis completes

📊Step 2: Review Your Audit Report

Once your repositories are analyzed, view your comprehensive audit report. This includes code health scores, risk predictions, technical debt analysis, and maintenance cost estimates.

• View overall code health score (0-100)
• See module-level risk predictions
• Review technical debt breakdown
• Check maintenance cost estimates
• Export data (JSON, CSV, PDF) for further analysis
• Create GitHub issues from critical findings
• Set alerts for health score changes
• Schedule recurring audits

🔮Step 3: Understand Predictions

Echeo uses machine learning models trained on 1000+ repositories to predict which modules will fail, which commits create debt, and how code health impacts business outcomes.

• Failure predictions with 85%+ accuracy
• Technical debt identification and prioritization
• Maintenance cost forecasting
• Team productivity insights
• Business outcome correlations

What is Code Health?

Code health is a composite metric (0-100) that predicts maintainability, reliability, and business outcomes. It combines multiple factors including complexity, test coverage, documentation quality, code churn, and historical patterns.

• 70+ - Excellent: Low maintenance costs, high reliability
• 50-69 - Good: Moderate maintenance, acceptable risk
• 30-49 - Fair: Higher maintenance costs, increased risk
• Below 30 - Poor: High maintenance costs, significant risk

Companies with code health scores above 70 raise 2.3x more funding than those below 50.

Key Metrics

Echeo tracks comprehensive metrics to assess code health:

• Complexity - Cyclomatic and cognitive complexity scores
• Test Coverage - Unit, integration, and E2E test coverage
• Documentation - Code comments, README quality, API docs
• Code Churn - Frequency of changes and refactoring patterns
• Technical Debt - Quantified debt ratio and maintenance costs
• Maintainability Index - Overall maintainability score

Reading Audit Reports

Audit reports provide detailed insights into your codebase health:

• Executive Summary - High-level code health score and key findings
• Module Analysis - Risk predictions for each module/component
• Commit Analysis - Which commits introduce technical debt
• Team Insights - Productivity metrics and code quality trends
• Recommendations - Prioritized actions to improve code health

Failure Predictions

Echeo predicts which modules are most likely to fail based on code patterns, complexity, and historical data. Our models achieve 85%+ accuracy by analyzing:

• Code complexity and coupling patterns
• Test coverage and quality
• Historical failure patterns in similar code
• Recent changes and churn rates
• Dependency health and vulnerabilities

Technical Debt Analysis

Technical debt is quantified and prioritized based on business impact:

• Debt Ratio - Percentage of codebase that needs refactoring
• Maintenance Cost - Estimated cost to maintain current code quality
• Refactoring Priority - Which modules should be refactored first
• ROI Analysis - Expected return on investment for refactoring

Business Impact

Code health directly impacts business outcomes:

• Funding Correlation - Companies with high code health raise 2.3x more funding
• AI Code Cost - AI-generated code costs 2.3x more to maintain
• Team Productivity - Higher code health = faster development velocity
• Technical Risk - Lower code health = higher risk of production issues

What is code health and how is it measured?

Code health is a composite metric (0-100) that predicts maintainability, reliability, and business outcomes. Echeo analyzes complexity, test coverage, documentation, code churn, and historical patterns to generate your score. Companies with scores above 70 raise 2.3x more funding than those below 50.

How accurate are the predictions?

Our failure prediction models achieve 85%+ accuracy by training on 1000+ repositories with known outcomes. We analyze code patterns, complexity, test coverage, and historical failure data to identify modules most likely to break. Predictions are continuously improved as we analyze more codebases.

Can I export my audit data?

Yes! You can export all audit data, metrics, and reports in multiple formats:

• JSON - For programmatic analysis and integration
• CSV - For spreadsheet analysis and data analysis
• PDF - For executive reports and presentations

Enterprise customers also have API access for real-time data integration with their analytics tools.

Is my code secure?

Absolutely. We follow a zero-training guarantee:

• We analyze code structure and patterns only (not full source code)
• Your actual source code never leaves your repositories
• All data is encrypted in transit and at rest
• You can disconnect your GitHub account at any time
• We never use your code to train AI models
• SOC2 compliant with regular security audits

How does code health impact business outcomes?

Research shows strong correlations between code health and business success:

• Funding - Companies with code health 70+ raise 2.3x more funding
• AI Code Cost - AI-generated code costs 2.3x more to maintain
• Productivity - Higher code health = faster development velocity
• Risk - Lower code health = higher risk of production failures
• Technical Debt - Quantified debt impacts maintenance costs

How do I improve my code health score?

To improve your code health score:

• Reduce code complexity through refactoring
• Increase test coverage (aim for 80%+)
• Improve documentation quality
• Address technical debt in high-priority modules
• Reduce code churn in critical paths
• Follow audit report recommendations