# Praetor

> Praetor is a single-prompt, multi-agent orchestration framework that transforms any codebase into a comprehensive production readiness and audit package. It dispatches 18 autonomous domain experts overseen by a 4-Judge Quality Council, performing multi-disciplinary audits (security, compliance, performance, accessibility, operations) with 100% source-code traceability. Runs inside long-context LLM environments (Claude, GPT-5, Gemini). No installation required.

## Documentation

- [README.md](README.md): Project overview, features, quick start guide, architecture, and agent roster.
- [prompt/GETTING_STARTED.md](prompt/GETTING_STARTED.md): Step-by-step first-time walkthrough with copy-paste examples.
- [prompt/VERSION.md](prompt/VERSION.md): Version history, design properties, and system capabilities.

## Architecture

- [wiki/Why-Praetor.md](wiki/Why-Praetor.md): The problem Praetor solves and the swarm + gated QC architecture.
- [wiki/The-18-Expert-Agents.md](wiki/The-18-Expert-Agents.md): Detailed descriptions of all 18 specialized agent personas.
- [wiki/The-4-Judge-Quality-Council.md](wiki/The-4-Judge-Quality-Council.md): How the 4 independent judges validate coverage, citations, clarity, and skip-validity.
- [wiki/The-7-Execution-Phases.md](wiki/The-7-Execution-Phases.md): Full pipeline from source resolution to release gate output.

## Core System (Prompt Kit)

- [prompt/00-orchestrator/MASTER_PROMPT.md](prompt/00-orchestrator/MASTER_PROMPT.md): The master prompt — paste this to start a Praetor run.
- [prompt/00-orchestrator/AGENT_ROSTER.md](prompt/00-orchestrator/AGENT_ROSTER.md): All 18 agents and QC judges at a glance.
- [prompt/07-agents/](prompt/07-agents/): Individual persona definitions for each of the 18 expert agents.
- [prompt/08-protocols/](prompt/08-protocols/): Inter-agent protocols — citations, handoffs, quality gates, conditional continue, resumable state.

## Reference

- [prompt/99-reference/GLOSSARY.md](prompt/99-reference/GLOSSARY.md): Plain-language definitions of all Praetor terminology.
- [prompt/99-reference/CHEATSHEET.md](prompt/99-reference/CHEATSHEET.md): Operator quick-reference for commands and overrides.
- [prompt/99-reference/ID_SCHEMES.md](prompt/99-reference/ID_SCHEMES.md): All 11 ID schemes used across artifacts and registers.
- [prompt/99-reference/BY_THE_NUMBERS.md](prompt/99-reference/BY_THE_NUMBERS.md): Canonical counts — phases, agents, registers, protocols.

## Community

- [CONTRIBUTING.md](CONTRIBUTING.md): How to contribute prompt content, mandate enhancements, and consistency fixes.
- [CODE_OF_CONDUCT.md](CODE_OF_CONDUCT.md): Community standards and expected behavior.
- [SECURITY.md](SECURITY.md): Vulnerability reporting and disclosure policy.
