Rigour

Paste this in the Content field:

# Rigour — Deterministic Quality Gates for AI Coding Agents

Rigour is an MCP server that forces AI agents to write production-grade code. It runs 23 automated quality checks on every file the agent writes, returning PASS/FAIL results that the agent must address before proceeding.

## Why Rigour?

AI coding agents (Claude, GPT, Copilot) generate code fast — but without quality enforcement. Rigour acts as a leash, not a suggestion. Every file gets scanned for:

- **Structural issues**: file size, cyclomatic complexity, deep nesting, long functions
- **Security vulnerabilities**: SQL injection, XSS, hardcoded secrets, command injection, path traversal
- **AI-drift patterns**: duplicated functions across files, hallucinated imports, context window quality degradation, inconsistent error handling
- **Agent governance**: retry loop breaking, checkpoint enforcement, environment alignment

## Two-Score System

Every scan produces:
- **AI Health Score** — measures AI-specific code quality (drift, hallucinations, async safety)
- **Structural Score** — measures traditional code quality (complexity, size, patterns)

Both scores use severity-weighted deductions: critical issues cost 20 points, high costs 10, medium costs 5.

## Quick Start

Add to your Claude Desktop or Cursor config:

```json
{
  "mcpServers": {
    "rigour": {
      "command": "npx",
      "args": ["-y", "@rigour-labs/mcp"]
    }
  }
}

Supported Languages

TypeScript, JavaScript, Python, Go, Ruby, C#

Links

• GitHub: https://github.com/rigour-labs/rigour
• npm: @rigour-labs/cli, @rigour-labs/core, @rigour-labs/mcp
• Website: https://rigour.run