DriftOps 🧠

AI-powered infrastructure compliance, drift detection, and auto-diagramming — as a single pipeline step.

---

What It Does

DriftOps drops into any CI/CD pipeline as a single step and gives your infrastructure a brain:

Stage	What happens
🔍 Scan	Reads your Terraform / ARM / Bicep / CDK on every push
📸 Snapshot	Captures infrastructure state, stores it
🔄 Diff	Compares to prior state — what changed, what drifted
🏛️ Comply	AI benchmarks every resource against NIST 800-53 + CIS
📊 Diagram	Auto-generates L1–L4 architecture diagrams
🚦 Gate	Blocks the deploy if critical violations found
💬 Report	Posts AI-written summary to PR as a comment

Quick Start

Add this to your GitHub Actions workflow:

- name: DriftOps Scan
  uses: driftops-dev/driftops@v1
  with:
    iac_path: './terraform'
    compliance_level: 'nist-800-53'
    enforce: false         # set true to block on violations
  env:
    DRIFTOPS_TOKEN: ${{ secrets.DRIFTOPS_TOKEN }}

Stack

• CI/CD: GitHub Actions / Azure DevOps
• Frontend: Cloudflare Pages (React)
• Backend: Cloudflare Workers
• Database: Supabase (Postgres)
• Auth: Supabase Auth
• AI: Groq (free tier) → Claude Haiku at scale
• SSL + Firewall: Cloudflare

Cost to run: $0/month at POC scale.

Project Structure

driftops/
├── action/          # GitHub Action + ADO Extension
├── dashboard/       # Cloudflare Pages frontend
├── worker/          # Cloudflare Worker API
├── supabase/        # Database schema + migrations
└── docs/            # Documentation

Roadmap

• Repo scaffold
• Terraform scanner
• State diff engine
• AI compliance mapping (NIST 800-53)
• GitHub Action packaging
• Dashboard (compliance score + drift timeline)
• Auto-diagram generation (L1–L4)
• Azure DevOps extension
• Enforce / auto-remediation mode
• CIS Benchmark support
• SOC2 mapping

License

MIT