Skip to content
Harery edited this page Jun 14, 2026 · 6 revisions

Welcome to the Praetor Wiki

Praetor is an autonomous multi-agent orchestration framework designed to audit codebases and generate production-readiness packages.

By utilizing 18 autonomous expert agents under the strict review of a 4-Judge Quality Council, Praetor scans repository sources to compile security, SRE runbooks, test specifications, business validation matrices, and compliance control maps (SOC2, GDPR, HIPAA)—all with 100% verified source-code citations (file:line).


📦 Installation

Method Command / Link Notes
npm npx praetor-audit-kit
Install to tools npx praetor-audit-kit --install Auto-detects 13 agentic tools (Claude Code, OpenCode, Cursor, Windsurf, GitHub Copilot, Cline, Amazon Q, Gemini CLI, Aider, Continue.dev, Codex CLI, Roo Code, Sourcegraph Cody)
GitHub Release Download v2.9.0
Scope selection npx praetor-audit-kit --scope biz Number, alias, or range
Uninstall npx praetor-audit-kit --uninstall Removes from all tools
Direct Copy prompt/00-orchestrator/MASTER_PROMPT.md into your LLM context

📖 Wiki Table of Contents

1. Introduction

  • Why Praetor?: Learn why traditional static code review tools and generic single-agent LLM reviews fail compared to a specialized multi-agent swarm.
  • Quick Start Guide: Get Praetor running on your codebase in less than 2 minutes.

2. Core Swarm Architecture

  • The 18 Expert Agents: Meet the specialized personas (DevOps, Security, WCAG, Chaos, SRE) that audit your codebase.
  • The 4-Judge Quality Council: Understand the zero-hallucination engine that validates all generated findings and citations.
  • The 7 Execution Phases: Walk through the automated pipeline from source resolution to final gate delivery.

3. Deliverables & Streams

4. Advanced Usage

Clone this wiki locally