randomittin/superx

3 stars · Last commit 2026-04-26

Autonomous Claude Code agent — 14 specialized agents, 10-parallel execution, project-specific skill learning, mechanical checkpointing, wave-based planning with acceptance gates, caveman ultra (~75% token savings). One prompt → finished project.

PluginWorkflow & Automation TDD & Testing Debugging Documentation Design & UI DevOps & Infra Productivity AI & Prompting Data & ML Security

README preview

# superx

**One prompt → finished project.** 14 specialized agents, 10-parallel execution, project-specific skill learning, mechanical checkpointing, ~75% token savings. Beats raw Claude Code usage.

[![Claude Code](https://img.shields.io/badge/Claude%20Code-plugin-e056a0?style=flat-square)](https://code.claude.com)
[![License: MIT](https://img.shields.io/badge/License-MIT-9b59b6?style=flat-square)](LICENSE)
[![Version](https://img.shields.io/badge/version-1.1.0-00d4ff?style=flat-square)](CHANGELOG.md)
[![Install](https://img.shields.io/badge/install-curl-4ecca3?style=flat-square)](#quick-start)

```bash
curl -fsSL https://raw.githubusercontent.com/randomittin/superx/main/install.sh | bash
```

<details>
<summary><strong>What is superx?</strong></summary>

superx makes Claude Code **dramatically more capable** than using it directly:

- **14 specialized agents** (security auditor, database architect, incident responder, etc.) — each with the right model + effort level
- **10-parallel execution** — independent files/projects run simultaneously, not sequentially

View full repository on GitHub →