planning-phase

Transform enriched user stories into an execution plan with parallel waves and agent assignments.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/metasaver/claude-marketplace --skill planning-phase-metasaver
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: planning-phase
Source: https://github.com/metasaver/claude-marketplace/tree/main/plugins/metasaver-core/skills/workflow-steps/planning-phase
Command: npx skills add https://github.com/metasaver/claude-marketplace --skill planning-phase-metasaver

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill converts enriched user stories into an execution plan with waves, dependencies, and agent assignments.

Core Features & Use Cases

  • Wave-based planning: Groups independent stories into execution waves.
  • Agent mapping: Assigns domain agents per story content.
  • Output plan: Produces a structured execution-plan.md with tasks and dependencies.

Quick Start

After architecture annotations, run planning-phase to generate the execution plan.

Frequently Asked Questions about planning-phase

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

FAQPage Schema
How do I convert user stories into an execution plan with parallel waves?

Planning-phase transforms enriched user stories into an execution plan by analyzing dependencies, grouping independent stories into parallel waves, and assigning domain agents. It produces a structured execution-plan.md with tasks, dependencies, and agent mappings for coordinated execution.

What dependencies do I need to analyze before creating execution waves?

Planning-phase reads 'Depends On' and 'parallelizable_with' annotations from your user stories to build a dependency graph. These enriched story annotations determine which stories can run in parallel and which must execute sequentially across waves.

How does planning-phase assign agents to execution tasks?

The Skill analyzes story content to map domain agents to each task and enforces a limit of 10 concurrent agents per wave. It spawns a project manager and outputs agent assignments alongside wave groupings in the execution plan.

Can I use planning-phase after architecture annotations?

Yes. Planning-phase is designed to run after architecture annotations have enriched your user stories. It consumes those annotations to determine dependencies and wave boundaries, then produces the final execution plan with agent assignments.

What output does planning-phase generate?

Planning-phase generates execution-plan.md containing structured wave data, task definitions, dependencies, and domain-agent mappings. The file provides a complete roadmap for executing user stories in parallel waves with assigned responsibilities.

Do I need to manually group stories by dependency before planning-phase runs?

No. Planning-phase automatically performs topological sorting to determine execution waves from your story dependencies. You only need to ensure stories are enriched with 'Depends On' and 'parallelizable_with' annotations.