Full reference guide for Ouroboros commands and agents
中文简介正在准备中。当前显示原文说明。
查看安装方法安装前可以查看交给 AI 代理的指令内容。
Full reference guide for Ouroboros power users.
ooo help
/ouroboros:ouroboros-help
A specification-first workflow engine: it turns vague ideas into a validated Seed (goal, constraints, acceptance criteria) before any code is written.
Interview → Seed → Execute → Evaluate
↑ ↓
└─── Evolutionary Loop ─────┘
Each ooo <name> is also the skill /ouroboros:<name>, except run, status,
config, and help, which are /ouroboros:ouroboros-<name>.
| Command | Purpose | Needs |
|---|---|---|
ooo / ooo welcome | First-touch onboarding | none |
ooo setup | Claude Code setup wizard | none |
ooo tutorial | Interactive hands-on learning | none |
ooo help | This reference | none |
ooo interview | Socratic interview that exposes hidden assumptions | MCP, in-session fallback |
ooo seed | Freeze interview results into a Seed | MCP, in-session fallback |
ooo auto | Goal to A-grade Seed, then execute and verify in the background | MCP |
ooo run | Execute a Seed and verify every acceptance criterion | MCP |
ooo evaluate | Mechanical checks, semantic review per criterion, consensus when uncertain | MCP, in-session fallback |
ooo evolve | Start or monitor an evolutionary loop across Seed generations | MCP |
ooo ralph | Server-owned loop until verified (ouroboros_ralph runs it as a background job) | MCP |
ooo unstuck | Five lateral thinking personas, single or debate | MCP |
ooo status | Session status and goal drift | MCP |
ooo qa | General-purpose QA verdict for any artifact | MCP, in-session fallback |
ooo pm | PM-focused interview and PRD | MCP |
ooo brownfield | Scan and manage brownfield repo defaults; find <topic> ranks registered repos | MCP |
ooo config | Settings GUI for per-stage agents and models | CLI |
ooo cancel | Cancel stuck or orphaned executions | CLI |
ooo resume-session | List in-flight sessions and re-attach commands | CLI |
ooo update | Check for updates and upgrade | CLI |
ooo publish | Publish a Seed as GitHub Issues | gh |
ooo maintain | Triage and resolve maintainer issues and PRs | gh |
MCP = the Ouroboros MCP server, which the Claude marketplace plugin or a
supported host setup (ouroboros setup --runtime <runtime>) provides.
In-session fallback = runs advisory-only through a bundled agent when the
MCP server is unavailable. CLI = drives the ouroboros command.
In Claude Code these phrases route to a command without the ooo prefix:
| Phrase | Command |
|---|---|
| "interview me", "clarify requirements", "socratic interview" | ooo interview |
| "crystallize", "generate seed", "create seed", "freeze requirements" | ooo seed |
| "ouroboros run", "execute seed", "run seed", "run workflow" | ooo run |
| "evaluate this", "3-stage check", "verify execution" | ooo evaluate |
| "evolve", "evolutionary loop", "iterate until converged" | ooo evolve |
| "ralph", "don't stop", "must complete", "until it works", "keep going" | ooo ralph |
| "think sideways", "i'm stuck", "break through", "lateral thinking" | ooo unstuck |
| "am I drifting?", "drift check", "session status" | ooo status |
| "write prd", "pm interview", "product requirements", "create prd" | ooo pm |
| "qa check", "quality check" | ooo qa |
| "cancel execution", "stop job", "kill stuck", "abort execution" | ooo cancel |
| "update ouroboros", "upgrade ouroboros" | ooo update |
| "brownfield defaults", "brownfield scan" | ooo brownfield |
| Agent | Purpose |
|---|---|
ouroboros:socratic-interviewer | Exposes hidden assumptions through questioning |
ouroboros:ontologist | Finds root problems vs symptoms |
ouroboros:seed-architect | Crystallizes requirements into a Seed |
ouroboros:evaluator | Verifies results against acceptance criteria |
ouroboros:qa-judge | Renders QA verdicts for ooo qa |
ouroboros:contrarian | "Are we solving the wrong problem?" |
ouroboros:hacker | "Make it work first, elegance later" |
ouroboros:simplifier | "Cut scope to absolute minimum" |
ouroboros:researcher | "Stop coding, start investigating" |
ouroboros:architect | "Question the foundation, redesign if needed" |
The last five are the ooo unstuck personas. Other files in
src/ouroboros/agents/ are internal pipeline roles.
End the final response with exactly one line ◆ <current state> → next: <recommended action>. Derive the state from live session state via ouroboros_session_status when available, otherwise from this skill's actual outcome. Never use a linear Step N of M footer; when the next step is a real choice, list 2-3 options.
还没有评价。欢迎分享你的使用感受。
无需登录即可发表。欢迎分享使用感受、提问，或说明在哪些环境下无法运行。
还没有评论。来写第一条吧。
Use when the work spans several files or replaces a framework, language, auth, storage, or API layer (rewrites and migrations included), and the user wants it to end up working and tested. Asks only what is missing, writes a spec with acceptance criteria, then runs and verifies it in the background. Not for questions, small edits, or an existing Seed (use run).
中文简介正在准备中。当前显示原文说明。
Scan and manage brownfield repository/worktree defaults for interviews
中文简介正在准备中。当前显示原文说明。
Cancel stuck or orphaned executions
中文简介正在准备中。当前显示原文说明。
Use when the user wants proof that finished work meets its acceptance criteria, not just that tests pass: 'verify it is actually done', 'check it against the spec', 'evaluate the run'. Runs mechanical checks, a semantic review per criterion, and multi-model consensus when uncertain. With execution.auto_evolve on (the default), a rejected result with a Seed continues into a Ralph fix loop; pass auto_evolve: false to verify only.
中文简介正在准备中。当前显示原文说明。
Start or monitor an evolutionary development loop
中文简介正在准备中。当前显示原文说明。
Use when the user wants to build or change something but the request is vague or open-ended: 'help me figure out what to build', 'I'm not sure what it should do', 'add X' with no details, or no clear definition of done. Runs a short Socratic interview that turns the idea into goals, constraints, and acceptance criteria before any code is written. Not for simple questions, small clear edits, or an existing Seed.
中文简介正在准备中。当前显示原文说明。