Ross ROSS = Recommend OSS · open-source software intelligence for agents

Q00/ouroboros

Agent OS: the agent gets smarter on its own. We just hold the line: the grading command and expected result never make it into the success contract we hand it. Interview-gated, staged evaluation, budgeted evolution loop. MCP server, 13 runtimes: Claude Code, Codex CLI, Gemini CLI, OpenCode, Copilot, Kiro and more. observed · 2026-08-28

github.com/Q00/ouroboros · homepage · Python · MIT (permissive) observed · 2026-08-28

Health v2 · maintenance only

78/100

  • Activity 99
  • Release rhythm 87
  • Longevity 16
How is this computed?

round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-03. Adoption (stars, forks) is never an input.

  • gap_med: 1
  • age_days: 232
  • days_rel: 10
  • days_push: 8
  • n_releases_24m: 118

Full methodology

Adoption not part of the score

5687 stars · 569 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Ouroboros is an 'Agent OS' that manages AI coding agent workflows through interview-gated requirement capture, staged evaluation, and a budgeted evolution loop, keeping grading criteria hidden from the agent's success contract. It ships as a CLI (pip install ouroboros-ai) and MCP server/plugin supporting 13 coding agent runtimes including Claude Code, Codex CLI, Gemini CLI, OpenCode, and GitHub Copilot.

Use cases

  • run AI coding agents in a verifiable spec-then-verify loop
  • hide grading assertions from the agent so it cannot game evaluation
  • turn a vague idea into a reviewed spec via an ambiguity-scoring interview
  • orchestrate the same workflow across Claude Code, Codex, Gemini CLI and other runtimes
  • evaluate and iterate on agent-generated code with staged verification
  • record replayable evidence of AI coding work

When to choose

  • you use coding agents like Claude Code or Codex and want trustworthy, spec-pinned verification of their output
  • you want an agent loop that improves across generations without leaking the success criteria
  • you need one orchestration engine across multiple agent runtimes

When to avoid

  • you need a simple one-shot code generator without evaluation overhead
  • you don't use any supported coding agent runtime
  • you want a lightweight prompt library rather than a full workflow engine

Facets

cli-tool · maturity active

agent-framework llm-inference mcp cli developer-tools workflow-automation testing large-language-models developer-tools cli python cross-platform windows agent-os llm-orchestration agentic-ai claude-code codex-cli gemini-cli opencode github-copilot llm-evaluation coding-agent loop-engineering spec-driven-development interview-gated mcp-server self-improving-agent ai-agents automation command-line linux macos

4 sources

Member repositories

RepositoryRoleHealth v2
Q00/ouroborosmain78

For agents

markdown · JSON · MCP: product_card(name="Q00/ouroboros")

Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem