Introduction

TIP

One sentence: CodyMaster is the operating system around your repo for AI-assisted engineering — memory, gates, dashboard, and skills — not a replacement for your IDE or model.

What CodyMaster is

• A Node.js CLI (cm / codymaster) that registers command groups for projects, tasks, dashboard, engineering helpers, and distro checks (src/cli/command-registry.ts).
• A project memory layer under .cm/ (continuity file, optional SQLite context.db, context bus, token budget).
• A Mission Control dashboard (Express) for kanban-style visibility over data in ~/.codymaster/kanban.json (src/dashboard.ts, src/data.ts).
• An MCP context server exposing search, URI resolve, bus read/write, budget checks, advisory JSON surfaces, and engineering-bridge tools (src/mcp-context-server.ts).
• A skills library — Markdown instruction packs in skills/*/SKILL.md that agents (Claude, Cursor, etc.) can follow.

What CodyMaster is not

• Not a hosted SaaS product in this repo — it runs on your machine with a local-first default stack.
• Not a single “magic model” — it orchestrates your tools, files, and habits.

Who it is for

• Teams and individuals who use AI coding agents daily and need structure: memory, checklists, gates, and repeatable workflows.
• Maintainers who want one place for continuity, learnings, and handoff between humans and agents.

Next steps

• How it works
• Installation
• System architecture