Skip to content

Kibertum/SENAR

Repository files navigation

SENAR — Supervised Engineering & Normative AI Regulation

Version License

Open methodology for human-AI collaborative software development.

senar.tech · Читать на русском

What is SENAR

SENAR is a practical engineering standard for teams where AI agents write code alongside human engineers. It defines roles, ceremonies, quality gates, and metrics — not abstract principles, but concrete rules you can adopt today. SENAR scales from a solo developer with an AI assistant to enterprise teams with dozens of agents. It is vendor-neutral and works with any AI coding tool: Cursor, Claude Code, Copilot, Windsurf, or custom setups.

Start Here

New to SENAR? Start with SENAR Core — 8 essential rules you can adopt in 5 minutes:

SENAR Core (English) | SENAR Core (Russian)

Then read the Quick Start Guide for a hands-on walkthrough.

PDF Downloads

Documentation

Document Description Link
Standard v1.3 Full specification: roles, rules, quality gates, metrics, conformance standard/
Guide Practical guide: philosophy, adoption, walkthroughs (EN + RU) guide/
Reference Reference materials: glossary, scaling ratios, governance, tooling reference/
Core v1.3 8 rules quick start for individual developers core/

Key Concepts

  • Context-over-Code — AI effectiveness depends on context quality, not prompt tricks
  • FPSR (First Pass Success Rate) — primary metric: percentage of tasks done right on first attempt
  • Quality Gates — automated checkpoints (QG-0 before work, QG-2 before merge)
  • 15 Rules — concrete engineering rules, not abstract principles
  • 4 Configurations — Core, Foundation, Team, Enterprise — adopt incrementally

Who Is This For

  • Engineering teams using AI coding assistants (Cursor, Claude Code, Copilot, etc.)
  • CTOs and engineering leaders building AI-native development processes
  • Organizations seeking a structured approach to human-AI collaboration
  • Solo developers who want reliable AI-assisted workflows

License

CC BY-SA 4.0 — use freely with attribution.

Authors: Andrey Yumashev, Vadim Soglaev

About

SENAR — Supervised Engineering & Normative AI Regulation. Open methodology for human-AI collaborative software development. 15 rules, 5 quality gates, 10 metrics. v1.3

Topics

Resources

License

Stars

Watchers

Forks

Packages

 
 
 

Contributors