# Coordinate Claude Code Agent Teams

Complex coding tasks are hard to split, verify, and resume across multiple agents. OMC adds team, autopilot, and verify/fix workflows for Claude Code projects.

## Install

```bash
npx skillstore add supercent-io/omc
```

## Metadata

- Status: approved
- Slug: supercent-io-omc
- Version: 4.2.15
- Author version: 4.2.15
- Skillstore revision: r1
- Version status: valid
- Tree hash: 9beddbcb9a0c84a3f6771c33bae5e83366be1e01b837298dfdfa3cdf1a5e654b
- Author: supercent-io
- GitHub username: supercent-io
- License: MIT
- Repository: https://github.com/supercent-io/skills-template/tree/main/.agent-skills/omc/
- Ref: f93e9bb0daca99badb6a7e574b97737155d57cb3
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: confirmation\_required
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: external\_commands, network, filesystem
- Quality score: 38
- Quality tier: warning
- Public page: https://skillstore.pages.dev/skills/supercent-io-omc
- Manifest: https://skillstore.pages.dev/api/skills/supercent-io-omc/manifest

## Capabilities

- Provides installation and setup steps for the oh-my-claudecode plugin.
- Documents team, autopilot, ralph, ultrawork, plan, swarm, and ultrapilot modes.
- Shows commands for team pipelines, updates, setup, and doctor diagnostics.
- Explains optional Gemini CLI and Codex CLI use for cross-validation.
- Describes notification callbacks for Telegram and Discord stop events.
- Lists native teams configuration for Claude Code settings.

## Use Cases

- Coordinate Feature Implementation: Use team mode to split a feature into planning, execution, verification, and fixes.
- Review Complex Refactors: Use optional Codex or Gemini checks to compare architecture and review feedback.
- Standardize Agent Workflows: Document setup, update, and diagnostic commands for teams using Claude Code.

## Prompt Templates

### Plan A Small Feature

```
Use OMC to plan a user profile edit feature. Ask me any needed questions before implementation.
```

### Run Team Execution

```
Use /omc:team 3:executor "fix the failing tests in this repository" and summarize each stage before changes.
```

### Verify A Refactor

```
Use ralph mode to refactor the authentication module, then run verification and list remaining risks.
```

### Cross-Check Architecture

```
Use OMC team mode with optional Codex review to design a migration plan for billing. Include assumptions, tests, and rollback steps.
```

## Limitations

- Requires installing a third-party Claude Code plugin before most commands work.
- Some modes run autonomous or persistent loops and need user supervision.
- Native teams require editing the Claude Code settings file in the user home directory.
- Optional external CLIs and notification webhooks require separate security review.

## Best Practices

- Review plugin source and permissions before running install or setup commands.
- Use scoped tasks with explicit files, tests, and stop conditions.
- Store tokens and webhooks in secure configuration, not shared prompts or logs.

## Anti Patterns

- Do not run autopilot or ralph on broad requests without human checkpoints.
- Do not install optional CLIs or plugins from unfamiliar sources during urgent work.
- Do not paste production secrets into notification setup prompts or shell history.

## Security Audit

- Audited at: 2026-07-07T05:04:47.459\+00:00
- Summary: Most static external-command alerts are Markdown backtick false positives, but several install, update, setup, daemon, and notification commands are real risks. The skill also pairs broad Bash and file-write permissions with autonomous and persistent execution modes. No prompt injection text was found in the reviewed files.

## Stats

- Views: 101
- Downloads: 14
- Favorites: 0
- Popularity score: 0