agile-pm

Automate sprint planning and task decomposition with one-task-per-file isolation.

1|Updated Mar 6, 2026
One-click install
npx skills add https://github.com/grant-vine/wunderkind --skill agile-pm
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agile-pm
Source: https://github.com/grant-vine/wunderkind/tree/main/skills/agile-pm
Command: npx skills add https://github.com/grant-vine/wunderkind --skill agile-pm

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Agile PM helps AI agents plan and structure work so tasks are decomposed into isolated, non-conflicting units that can be executed in parallel.

Core Features & Use Cases

  • One-task-per-file: Enforces strict isolation of concerns to prevent file conflicts when multiple agents operate concurrently.
  • Explicit delegation patterns: Uses task() calls to direct work to subagents with clear responsibilities and guardrails.
  • Sprint planning and backlog management: Provides structured backlogs, velocity estimates, and dependency-aware break downs for product teams.

Quick Start

Provide a backlog and let Agile PM generate a conflict-free, parallel-friendly task decomposition.

Frequently Asked Questions about agile-pm

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I prevent file conflicts when running parallel AI agents?

To prevent file conflicts during parallel execution, enforce one-task-per-file isolation. This strict separation of concerns ensures multiple AI agents can operate concurrently on the codebase without overwriting each other's work.

What is task decomposition for parallel AI workflows?

Task decomposition for parallel workflows is breaking down a product backlog into isolated, non-conflicting units. It uses explicit delegation patterns like task() calls to direct subagents with clear responsibilities and guardrails for concurrent execution.

How do I automate sprint planning and manage an engineering backlog?

Automate sprint planning by providing a backlog to generate a conflict-free, parallel-friendly breakdown. This creates structured backlogs with velocity estimates and dependency-aware sequencing for product and engineering teams.

How does dependency-aware sequencing work for parallel task execution?

Dependency-aware sequencing schedules tasks by respecting prerequisite chains while maximizing parallel work. It guarantees safe parallel execution by ensuring isolated tasks only run when their dependencies are fully resolved.

Can I use explicit task delegation syntax to direct AI subagents?

Yes, you can use explicit task delegation syntax like task() calls to direct work to subagents. This method assigns clear responsibilities and guardrails, ensuring work is traceable and executed safely in parallel.

What are the limitations of one-task-per-file isolation for sprint planning?

The limitation of one-task-per-file isolation is that it requires strict adherence to the decomposition structure. Complex, tightly-coupled engineering tasks may resist single-file isolation, making safe parallel execution difficult without careful delegation.