Claude Skill MartBrowse skillsWhat is a Skill?
 
← Back to list

Teammode

Orchestrate a named team of cooperating Codex workers under one leader session, with durable on-disk team state.

AutomationAdvanced★ 67,531⑂ 5,507AI score 7/10Last updated: Aug 9, 2026

What it does

Teammode turns several Codex workers into a coordinated team run by your main session as leader.

  • Transport selection first: inspects the active tool surface (including tool_search for deferred tools) and announces whether it will run on native MultiAgentV2 agents, Codex App threads, or fall back to a plain-subagent split.
  • Durable state via script: a bundled dependency-free Node script (team.mjs) writes .omo/teams/{session_id}/ with team.json, an auto-generated member manual guide.md, and a shared artifacts/ folder, all behind a per-team lock.
  • Composition by ownership, not job title: each member owns one concrete slice — a code area, an ownership boundary, or a perspective — so responsibilities never overlap.
  • Worktree isolation and integration: colliding members get their own git worktree; integrate merges member branches back with a merge commit.
  • Enforced teardown: archive every member, then delete team state so no zombie teams leak.

Who it's for

  • Codex users running large refactors or investigations across multiple parallel threads/agents.
  • Anyone with a decision-complete plan in .omo/plans/ who wants to execute its waves in parallel.
  • Teams that need file-level isolation between concurrent agents.
  • Not for you if the work is perfectly isolated or the goal is still ambiguous — the skill itself tells you to use plain subagents then.

Example uses

  1. Deep bug hunt: "Investigate this slow shutdown as a team" — the leader spawns app-server-lifecycle and mailbox-delivery members, relays findings between them, and synthesizes the diagnosis.
  2. Parallel plan execution: map the three independent todos of wave 1 in .omo/plans/auth-refactor.md to three members, give worktrees to the ones touching shared files, then integrate after all report.
  3. Clean shutdown: notify members, copy evidence into artifacts/, archive, worktree-remove, and delete the team state.

· · · Install guide · · ·

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the packages/omo-codex/plugin/components/teammode/skills/teammode folder from the GitHub repo code-yeongyu/oh-my-openagent into my ~/.claude/skills/teammode/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/code-yeongyu/oh-my-openagent.git && mkdir -p ~/.claude/skills && cp -r oh-my-openagent/packages/omo-codex/plugin/components/teammode/skills/teammode ~/.claude/skills/teammode

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal and go to your working directory.
  2. Clone the repository: git clone https://github.com/code-yeongyu/oh-my-openagent.git
  3. Create the skills folder: mkdir -p ~/.claude/skills
  4. Copy the skill: cp -r oh-my-openagent/packages/omo-codex/plugin/components/teammode/skills/teammode ~/.claude/skills/teammode
  5. Confirm that scripts/team.mjs came along — without it no team state can be created.
  6. Make sure Node.js (or Bun) is installed: node -v
  7. Restart your Claude Code / Codex session and ask something like "run this as a team" to trigger the skill.
  8. Note: real teams require Codex's MultiAgentV2 or codex_app.* tools; otherwise the skill will report the limitation and continue serially.
View source on GitHub ↗License: NOASSERTION