skills.rest
sumik5
sumik5COMMUNITY
Explore all skills by this author →
𝕏@shivasek5GitHub

agent-manager

Decompose high-level project strategies into parallelizable developer tasks.

2|2|Updated Feb 9, 2017
Search Tags:#project management#software development#dependency management#parallel execution#agent orchestration#worktree#task allocation
One-click install
npx skills add https://github.com/sumik5/dotfiles --skill agent-manager
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-manager
Source: https://github.com/sumik5/dotfiles/tree/main/claude-code/skills/agent-manager
Command: npx skills add https://github.com/sumik5/dotfiles --skill agent-manager

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill automates the complex and time-consuming process of breaking down high-level project strategies into actionable, parallelizable tasks for development teams. It eliminates manual task allocation, dependency mapping, and execution planning, ensuring efficient project flow and reducing management overhead.

Core Features & Use Cases

  • Strategic Task Decomposition: Transforms high-level product owner (PO) strategies into detailed, executable tasks for multiple developer agents.
  • Dependency & Parallelism Management: Analyzes task dependencies and identifies opportunities for parallel execution, maximizing development speed.
  • Worktree Information Relay: Ensures all developer agents work within the correct Git worktree, maintaining project integrity.
  • Use Case: When a PO defines a new feature like "Implement User Authentication," this Skill automatically creates a detailed plan, assigning frontend, backend, testing, and documentation tasks to individual developers, specifying their worktrees and execution order (e.g., parallel).

Quick Start

Instruct the Manager Agent to analyze the PO's strategy for "Implement User Authentication" and generate a task distribution plan for the Developer Agents.