Happy slice of mcp pizzaMCP.pizza

ouroboros

MCP.Pizza Chef: Q00

Instead of typing a rough request and hoping your AI coding assistant gets it right, this tool walks you through a short back-and-forth interview to nail down what you actually want, then locks that plan in place before any code gets written. It builds the project step by step and runs automatic checks afterward, so the assistant does not just say the work looks fine — it verifies it. It works inside terminal-based AI coding tools including Claude Code, Codex CLI, GitHub Copilot CLI, Gemini CLI, and several others, and needs no API key of its own; it simply directs whichever coding assistant you already have set up.

Coding

Use This MCP client To

Turn a rough app idea into a working, checked codebase Have my AI assistant ask clarifying questions before it starts coding Lock in a plan so my assistant does not wander off track mid-project Get an automatic check that the finished code actually works Keep a record of every decision made while my assistant built the project Reuse the same coding workflow across different AI assistants Hand off a coding task with a clear specification instead of a vague ask

README

English | 한국어 | 简体中文


◯ ─────────── ◯

Ouroboros

O U R O B O R O S

◯ ─────────── ◯

Stop prompting. Start specifying.
The Agent OS for replayable, specification-first AI coding workflows

PyPI Tests License GitHub Sponsors

Quick Start · Why · Results · How It Works · Commands · Philosophy

Turn a vague idea into a verified, working codebase -- across Claude Code, Codex CLI, OpenCode, Hermes, Gemini, Kiro, Copilot, Pi, and Zcode.

Ouroboros is an Agent OS for AI coding: a local-first runtime layer that turns non-deterministic agent work into a replayable, observable, policy-bound execution contract. It replaces ad-hoc prompting with a structured specification-first workflow: interview, crystallize, execute, evaluate, evolve.

The Ouroboros Agent OS Stack

Like any OS, Ouroboros is split into a stable OS layer of primitives, an application layer of domain workflows, and a shell that humans actually sit in front of. Three repos, one stack:

Layer Repo Role What it gives you
Shell (terminal client) Q00/ourocode Native terminal UI for running ooo workflows across Claude / Codex / Gemini CLIs in one session TUI, wonderTool decision pickers, MCP pane state, command discovery
Apps (domain workflows) Q00/ouroboros-plugins UserLevel plugin contract — composes core primitives into installable domain programs (PR ops, Jira sync, incidents, releases) Plugin manifest, scoped permissions, audit/provenance, reference plugins
OS (this repo) Q00/ouroboros Agent OS core — Seed, Ledger, Runtime, MCP, safety boundaries ooo commands, spec-first workflow engine, multi-runtime adapter

How they connect:

  ourocode  ──►  ooo / ouroboros-plugins  ──►  ouroboros core (Seed · Ledger · MCP · Runtime)
   shell             user-level apps                        kernel
  • The kernel (ouroboros) owns the contract: every action becomes a Seed-bound, ledger-recorded, replayable event — regardless of which LLM executes it.
  • Plugins (ouroboros-plugins) declare scoped capabilities against that contract, so domain workflows (review a PR, triage a Linear ticket, run a release) stay auditable and policy-bound instead of being one-off prompts.
  • Ourocode is the terminal shell: it surfaces MCP state, interview questions, and wonderTool decisions as first-class TUI elements, so you can drive the OS without leaving the keyboard or switching between CLIs.

Use ouroboros alone with any supported CLI, layer plugins on for domain workflows, or install ourocode when you want a unified terminal cockpit.

ouroboros FAQ

Can I use this to turn a vague project idea into working code?
Yes — it interviews you about what you want first, writes that down as a plan, then builds and checks the code against it.
Can I use this to catch mistakes before I see the finished code?
Yes — it runs an automatic evaluation step after building, rather than just declaring the work done.
Which AI coding assistants does this work with?
It works inside terminal-based coding tools such as Claude Code, Codex CLI, GitHub Copilot CLI, Gemini CLI, and a handful of others — you keep using the assistant you already have.
Do I need an account or an API key to use this?
No separate account or key is required for the tool itself — it runs on top of whichever AI assistant you have already set up, such as Claude, ChatGPT/Codex, or Gemini.
How hard is this to set up?
It requires comfort with a terminal — installation is a command-line step, so this is best suited to someone who already writes or runs code, not a fully non-technical user.
Do I need to know how to code to use this?
You do not need to write the code yourself, but getting it installed and running does require basic command-line use.
Will it just start changing my project without asking?
No — if it needs to make a one-time setup change, it asks for your go-ahead first.
Can I choose which AI model handles each step?
Yes — a configuration screen lets you pin a specific model to a stage of the process, though by default it follows your coding assistant's own settings.