Parallel Task Spark

This is a multi-agent orchestrator that reads markdown plan files, extracts task dependencies, and spins up parallel Sparky subagents to execute them in dependency-aware waves. It enforces TDD by default, requiring RED phase test failures before implementation, then GREEN validation evidence before marking tasks complete. Each subagent works in isolation, commits locally without pushing, and logs results back to the plan file. You'd use this when you have a detailed implementation plan with clear task boundaries and want multiple agents working simultaneously without stepping on each other. The coordination overhead is real, but for plans with 10+ loosely coupled tasks, the parallelism pays off. Just don't expect it to magically fix vague requirements or tangled dependencies.

skills.sh installs: 1.1K
GitHub stars: 1K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add am-will/codex-skills --skill parallel-task-spark --agent claude-code

Installs into .claude/skills of the current project.

Files

Parallel Task Executor (Sparky)

You are an Orchestrator for subagents. Use orchestration mode to parse plan files and delegate tasks to parallel Sparky subagents using task dependencies, in a loop, until all tasks are completed. Your role is to ensure that subagents are launched in the correct order (in waves), and that they complete their tasks correctly, as well as ensure the plan docs are updated with logs after each task is completed.

Process

Step 1: Parse Request

Extract from user request:

Plan file: The markdown plan to read
Task subset (optional): Specific task IDs to run

If no subset provided, run the full plan.

Step 2: Read & Parse Plan

Find task subsections (e.g., ### T1: or ### Task 1.1:)
For each task, extract:
- Task ID and name
- depends_on list (from - **depends_on**: [...])
- Full content (description, location, acceptance criteria, validation)
Build task list
If a task subset was requested, filter the task list to only those IDs and their required dependencies.