Skip to content

Quick Start

Navigate to your repository and initialize Certify:

Terminal window
cd your-repo
certify init

This creates:

.certification/
├── config.yml # configuration
├── policies/
│ ├── go-standard.yml # universal Go rules
│ └── go-library.yml # library-specific (auto-detected)

It also generates GitHub Actions workflows in .github/workflows/.

Scan your repository to find all certifiable code units:

Terminal window
certify scan

Output:

Detecting languages...
• go: 47 files (config found)
Go adapter: 182 symbols
✓ Discovered 195 code units (saved to .certification/index.json)

Certify finds functions, methods, types, and files using language-aware adapters.

Evaluate every unit against your policies:

Terminal window
certify certify

This collects evidence (lint results, test status, git history, AST structural analysis, complexity metrics), evaluates against policy packs, scores across 9 quality dimensions, and assigns certification status.

Collected 4 repo-level evidence items
Processing... 195/195
✓ Certified 195/195 units

Generate your report card:

Terminal window
certify report
╔══════════════════════════════════════════════════════════════╗
║ CERTIFY — REPORT CARD ║
╠══════════════════════════════════════════════════════════════╣
║ Overall Grade: 🟢 B+ Score: 87.3% ║
║ Total Units: 195 Pass Rate: 100.0% ║
╚══════════════════════════════════════════════════════════════╝

For the complete markdown report card:

Terminal window
certify report --format full

For a browsable static HTML site:

Terminal window
certify report --site

If you have an AI provider configured, run a 6-phase architectural analysis:

Terminal window
certify architect

This analyzes your package structure, dependencies, quality patterns, test strategy, and security posture — producing a comprehensive ARCHITECT_REVIEW.md with prioritized recommendations and projected metric improvements.

If anything went wrong, run the built-in diagnostics:

Terminal window
certify doctor

This checks your environment, configuration, policies, tools, AI providers, and analysis tiers per language — showing exactly what’s available and what to install for deeper analysis. See Doctor & Onboard →