ralph-parallel

Parallel spec execution plugin for Claude Code. Works with ralph-specum to dispatch spec tasks to Agent Teams for concurrent execution.

What It Does

Takes a spec generated by ralph-specum, analyzes task dependencies and file ownership, then dispatches parallelizable groups to Agent Team teammates for simultaneous execution. Coordinates the team through phase gates and quality checkpoints.

Install

/install-plugin kavanaghpatrick/ralph-parallel

Requires ralph-specum to be installed first.

Commands

Command	Description
/ralph-parallel:dispatch	Analyze tasks, partition into groups, spawn teammates
/ralph-parallel:status	Monitor parallel execution progress
/ralph-parallel:merge	Verify and integrate results (worktree strategy)

Strategies

• file-ownership (default): Each teammate owns specific files. No merge step needed.
• worktree: Each teammate gets a git worktree. Requires /merge after completion.

How It Works

1. Parse: Reads tasks.md from your spec, builds a dependency graph
2. Partition: Groups tasks by file ownership to avoid conflicts
3. Dispatch: Creates an Agent Team, spawns teammates with scoped prompts
4. Coordinate: Manages phase gates, quality checks, and teammate lifecycle
5. Complete: Shuts down team, marks dispatch as merged

Demo Results

• user-auth (12 tasks, 4 teammates): ~3.4x speedup
• gpu-graphics-demo (12 tasks, 3 teammates): Full Rust/wgpu app built in parallel

Requirements

• Claude Code with Agent Teams support
• ralph-specum plugin (for spec generation)
• Python 3.10+ (for analysis scripts)