Skip to main content

Welcome to Rigour

The Engineering Handshake for AI Agents 🎯

Rigour is a high-fidelity code review and supervision platform designed specifically for the age of agentic coding. It acts as the "bridge" between the creative power of AI models and the strict engineering requirements of production-grade codebases.

The Core Problem​

AI agents can refactor 1,000 lines of code in seconds, but they often struggle with:

  • Project Context: Understanding specific architectural boundaries.
  • Complexity Rails: Keeping functions small and manageable.
  • Safety: Avoiding modifications to critical CI/CD or infrastructure files.

Rigour solves this by providing a standardized "Technical Contract" that agents must satisfy before their work is accepted.

Why Rigour?​

  • 🛡️ Universal SME: Acting as a Subject Matter Expert for 10+ languages (Go, Rust, Java, Python, C++, and more).
  • 🧠 Architectural Gates: Enforces SOLID, DRY, and specialized paradigms (OOP, Functional).
  • 🔄 Quality Handshake: Bridges the gap between Static structure (SAST) and Dynamic performance (DAST/IAST).
  • 🤝 Agentic Handshake: The definitive protocol for reliable AI agent integration.

Two Modes of Operation​

Rigour is designed to be used by both Humans (who define the rules) and Agents (who follow them):

  1. Advisor Mode (MCP): For interactive development. The agent proactively asks Rigour for feedback during its task.
  2. Supervisor Mode (CLI): For headless automation. Rigour wraps the agent and strictly enforces a PASS status before finishing.

Learn about Operating Modes →

🌟 What's New in v2.0.0 (The Universal SME Release)​

Rigour v2.0.0 is a complete reimagining of the platform, moving from a TypeScript-focused linter to a Universal Engineering Supervisor.

  • Universal AST Support: High-fidelity structural parsing for 10+ languages (Go, Rust, Java, Python, C++, etc.).
  • Cognitive Complexity Engine: Nesting-aware complexity measurement.
  • Dynamic Quality Handshake: Requiring high test coverage for complex/high-risk code paths.
  • Interactive Human Reporting: The new --interactive flag for rich terminal visualizations.

Technical Book Structure​

This documentation is organized as a technical guide to help you ship faster and safer:

Installation​

# Register globally
npm install -g @rigour-labs/cli

# Run your first check
rigour check --interactive