Ouroboros

Active
GitHub Python MIT

Description

Ouroboros is a spec-driven multi-agent framework that shifts from traditional prompting to specification-driven development, supporting multi-agent collaboration, MCP tool integration, and automated workflow orchestration for building high-quality agent systems.

Key Features

  • Spec-first workflow: Socratic interview → immutable seed spec → execution → evaluation
  • Multi-runtime adapter supporting Claude Code, Codex CLI, OpenCode, Hermes, Gemini, Kiro, Copilot
  • Three-layer architecture: Kernel (contract engine), Plugins (domain workflows), Shell (TUI cockpit)
  • Ledger-recorded, replayable execution events for audit and observability
  • MCP integration for tool discovery and scoped permissions
  • Plugin ecosystem with composable domain programs (PR ops, Jira sync, incidents, releases)

Use Cases

💡 Structured AI coding workflows replacing ad-hoc prompting
💡 Replayable and auditable agent execution for team environments
💡 Multi-agent collaboration with policy-bound tool access
💡 Domain-specific automation via installable plugins

Tags

multi-agent framework agent python mcp workflow automation

Categories

🧠 Multi-Agent 🤖 Agent Framework

Quick Start

Install with `curl -fsSL https://raw.githubusercontent.com/Q00/ouroboros/main/scripts/install.sh | bash`. Then open your AI coding agent and run `ooo interview "I want to build a task management CLI"`. The installer auto-detects Claude Code, Codex CLI, and Hermes CLI. For other runtimes, run `ouroboros setup --runtime <opencode|kiro|copilot|gemini|pi>`.
Visit GitHub

Related Projects