Ouroboros

listing.by

Run a spec-first workflow for AI coding agents: interview, lock a seed spec, execute, evaluate and evolve.

state.local.long (state.local.title)listing.badge.runs_with

listing.install

listing.local.explain

uvx ouroboros-ai

listing.local.needs listing.local.note

act.add_to a11y.new_tabact.add_to a11y.new_tab

listing.share.h

listing.share.p_local

mcp.tc/i/ouroboros

listing.about

Ouroboros is an Agent OS for AI coding. It turns a vague request into a structured, immutable seed spec through a Socratic interview. It then executes against that spec and checks the result with a staged evaluation gate. Actions are recorded in a ledger so runs can be replayed. The grading command and expected output are kept out of the success contract handed to the agent.

It runs locally as the Python package ouroboros-ai, started with uvx over stdio. It works with several coding agent hosts, including Claude Code, Codex CLI, Gemini CLI, OpenCode, Copilot and Kiro. No API key is declared for the server itself, though the host agent needs its own model access. An install script is also provided for macOS, Linux, WSL 2 and Windows.

listing.can

  • Run a Socratic interview that exposes hidden assumptions in a request
  • Lock intent into an immutable seed spec before code is written
  • Execute work against the spec through a supported coding agent
  • Evaluate results with a 3-stage automated gate
  • Run an evolution loop with a budget
  • Replay and audit runs through a ledger
  • Work across many agent runtimes from one setup

listing.tools

listing.notools.local

listing.prompts

  • ooo interview "I want to build a task management CLI"

  • Run ooo setup to configure Ouroboros for this project

  • Interview me about this feature idea and produce a seed spec

  • Evaluate the current implementation against the seed spec

listing.setup.h

listing.setup.local

Claude Code

  1. Run this in a terminal, in your project folder:
claude mcp add --transport stdio ouroboros -- uvx ouroboros-ai
  1. Start Claude Code and type /mcp. ouroboros should show as connected.

Add --scope user to make it available in every project.

Claude Desktop

  1. Open Settings → Developer → Edit Config. It opens claude_desktop_config.json. Add:
claude_desktop_config.json
{
  "mcpServers": {
    "ouroboros": {
      "command": "uvx",
      "args": [
        "ouroboros-ai"
      ]
    }
  }
}
  1. Save the file and restart Claude Desktop.

Needs uv (Python) on your computer. The file lives in ~/Library/Application Support/Claude/ on macOS and %APPDATA%\Claude\ on Windows.

Cursor

Add to Cursor a11y.new_tab

Or add it by hand to ~/.cursor/mcp.json (all projects) or .cursor/mcp.json (this project):

mcp.json
{
  "mcpServers": {
    "ouroboros": {
      "command": "uvx",
      "args": [
        "ouroboros-ai"
      ]
    }
  }
}

Needs uv (Python) on your computer.

VS Code

Install in VS Code a11y.new_tab

Or from a terminal:

code --add-mcp '{"name":"ouroboros","type":"stdio","command":"uvx","args":["ouroboros-ai"]}'

Or commit it to the repo in .vscode/mcp.json:

.vscode/mcp.json
{
  "servers": {
    "ouroboros": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "ouroboros-ai"
      ]
    }
  }
}

Needs uv (Python) on your computer.

Devin Desktop

  1. Add it to ~/.config/devin/mcp_config.json (macOS and Linux) or %APPDATA%\devin\mcp_config.json (Windows):
mcp_config.json
{
  "mcpServers": {
    "ouroboros": {
      "command": "uvx",
      "args": [
        "ouroboros-ai"
      ]
    }
  }
}
  1. Refresh the MCP server list in Cascade.

Devin Desktop is the new name for Windsurf.

Codex

codex mcp add ouroboros -- uvx ouroboros-ai

Or edit ~/.codex/config.toml directly:

config.toml
[mcp_servers.ouroboros]
command = "uvx"
args = ["ouroboros-ai"]

Needs uv (Python) on your computer.

Gemini CLI

gemini mcp add ouroboros uvx ouroboros-ai

This adds it to the current project. Add -s user to use it everywhere.

Any client

Most clients that start local servers accept this shape:

{
  "mcpServers": {
    "ouroboros": {
      "command": "uvx",
      "args": [
        "ouroboros-ai"
      ]
    }
  }
}

Zed puts servers under context_servers in its settings, with the same command, args and env fields.

Needs uv (Python) on your computer.

listing.faq

Can I paste mcp.tc/i/ouroboros into my MCP client?

No. Ouroboros runs on your own computer, started by your client, so it has no web address to connect to. The quick link is the page to share; the install command is uvx ouroboros-ai.

Does Ouroboros need an API key?

No. It doesn't declare any keys or environment variables. It runs with your user account's permissions, so check what its tools can reach.

Is Ouroboros a remote or a local server?

Local. Your client starts it as a process on your computer with uvx ouroboros-ai, which needs uv (Python).

What can Ouroboros do?

You can run a Socratic interview that exposes hidden assumptions in a request, lock intent into an immutable seed spec before code is written and execute work against the spec through a supported coding agent.

Which clients can use it?

Any client that starts local servers: Claude Code, Claude Desktop, Cursor, VS Code, Devin Desktop, Codex, Gemini CLI, Zed and others. claude.ai and ChatGPT only connect to remote servers.

Who wrote this page?

mcp.tc's robot read Ouroboros's public metadata (its package and repository pages), and an AI model drafted the text from it. A person reviews anything the checks can't confirm. The text can still be wrong, so if you spot a mistake, use Report this listing on this page.

listing.em.h

listing.em.lead_local listing.em.how

listing.em.badge

em.prev_lightem.prev_dark
Markdown
[![Ouroboros on mcp.tc](https://mcp.tc/i/ouroboros/badge.svg)](https://mcp.tc/i/ouroboros)
HTML
<a href="https://mcp.tc/i/ouroboros"><img src="https://mcp.tc/i/ouroboros/badge.svg" alt="Ouroboros on mcp.tc" height="20"></a>

listing.em.badge_note

listing.em.widget

HTML
<script src="https://mcp.tc/w/ouroboros.js" async></script>

listing.em.widget_note

iframe

HTML
<iframe src="https://mcp.tc/embed/ouroboros" title="Ouroboros on mcp.tc" width="420" height="200" loading="lazy" allow="clipboard-write" style="border:0;border-radius:8px;max-width:100%"></iframe>

listing.em.iframe_note

JSON

listing.em.json_local

https://mcp.tc/i/ouroboros.json

listing.em.fields

listing.related

listing.more_in

  • Repowise

    repowise-dev

    state.local.label (state.local.title)

    Index a codebase once and give coding agents cited answers, change impact, git history, decisions and code-health data.

    mcp.tc/i/repowise

  • Repomix

    yamadashy

    state.local.label (state.local.title)

    Pack local or remote codebases into AI-friendly files that LLMs and coding agents can read or search.

    mcp.tc/i/repomix

  • .FAF Context

    Wolfe-Jam

    state.local.label (state.local.title)

    Keep one project.faf file in your repo and write AGENTS.md, .cursorrules, GEMINI.md and CLAUDE.md from it, with a context score.

    mcp.tc/i/faf-context

  • Nauro

    Nauro-AI

    state.local.label (state.local.title)

    Keep approved project decisions and their reasoning in a markdown store that connected coding agents read and update.

    mcp.tc/i/nauro

  • IFC MCP

    flinker-app

    state.local.label (state.local.title)

    Inspect and create IFC/BIM files, build reports and view models locally with IfcOpenShell running in Pyodide.

    mcp.tc/i/ifc

  • Kotlin & Java Library Sources

    aoreshkov

    state.local.label (state.local.title)

    Read the real sources, public API and KDoc of any Maven-published Kotlin or Java library from its sources jar.

    mcp.tc/i/kotlin-java-library-sources

listing.disclaimer_checked