一键导入
tutorial
Interactive guided tour of omg — progressive walkthrough from discovery to full orchestration
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Interactive guided tour of omg — progressive walkthrough from discovery to full orchestration
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Detect and clean AI-typical writing patterns in documents — em dashes, filler phrases, over-formality
Route questions to Claude, Codex, or Gemini for multi-perspective answers
Claude-Codex-Gemini tri-model orchestration — decompose, query, synthesize
Health check — verifies omg plugin is correctly installed and working
Quick reference — what omg agents and skills are available and how to use them
About omg — version, architecture, capabilities, and credits
| name | tutorial |
| description | Interactive guided tour of omg — progressive walkthrough from discovery to full orchestration |
| tags | ["discovery","meta","onboarding"] |
Guide the user through 5 progressive steps. After each step, explain what happened under the hood and ask if they want to continue.
Say to the user:
Welcome to omg! You have 28 specialized AI agents and 40 skills at your disposal. Let's start by meeting the team.
Try this:
about omgThis shows your agent roster, architecture, and what Copilot capabilities omg uses.
What happens: The about skill activates and displays the omg overview.
Explain: "omg has two types of agents: Orchestrators (9) that coordinate work, and Specialists (19) that do focused tasks. The orchestrators delegate to specialists — like a project manager coordinating a team."
Say to the user:
Now let's try a specialist. The explore agent maps codebases fast.
Try this:
deepsearch: find all configuration files in this project
What happens: The deepsearch skill activates, the explore agent searches from multiple angles.
Explain: "The explore agent used 5+ search strategies (exact name, glob patterns, imports, tests, docs) and cross-validated findings. This is more thorough than a single grep."
Say to the user:
Before building anything, omg can help you plan. The plan skill creates structured plans with testable acceptance criteria.
Try this:
plan: add a health check endpoint to this project
What happens: The plan skill activates. The agent interviews the user, explores the codebase, and creates a structured plan.
Explain: "The plan includes acceptance criteria that are testable (pass/fail, not subjective). When you're ready to execute, omg agents can use this plan as input — plans persist in .omg/plans/ across sessions."
Say to the user:
Now the main event — let omg's autopilot orchestrate a full implementation.
Try this (on a small task to see it in action):
autopilot: add a --version flag to the CLI that prints the package version
What happens: The autopilot agent coordinates the full lifecycle:
Explain: "Autopilot is one of 9 orchestrator agents. It coordinates specialists — it reads code (via explore), implements (via executor), and verifies (via verifier). Everything is persisted to .omg/ for cross-session continuity."
Say to the user:
You've seen the basics. Here are the power workflows:
Command What it does ralph: fix all lint errorsKeeps working until EVERY error is fixed with proof team 3: add tests for src/3 agents work in parallel on independent files trace: why did the build break?Competing hypotheses ranked by evidence deep interview: I want to build...Socratic Q&A → crystallizes vague ideas into specs research-to-pr: fix auth bugInvestigates → cloud agent creates PR automatically tdd: add email validationStrict red-green-refactor discipline For the full reference: say "help" anytime.
Power tip (interactive mode only):
/fleet review src/ for security, code quality, and architecture in parallel
/fleetdispatches multiple agents simultaneously — like having 3 reviewers working at once.
Remind the user:
.omg/ — resume anytimetutorial, show me how, getting started
copilot -i "tutorial"