delivery-plan

Convert a Task Packet into an ordered Delivery Plan with dependencies.

Updated Apr 17, 2026
One-click install
npx skills add https://github.com/KingSharkG/ai-agents-workflow --skill delivery-plan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: delivery-plan
Source: https://github.com/KingSharkG/ai-agents-workflow/tree/main/skills/delivery-plan
Command: npx skills add https://github.com/KingSharkG/ai-agents-workflow --skill delivery-plan

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Convert a Task Packet into an ordered Delivery Plan with dependencies and a Definition of Done, enabling Delivery PMs to orchestrate execution with clear, verifiable steps.

Core Features & Use Cases

  • Generate an ordered plan from a Task Packet, splitting multi-domain work into single-domain subtasks.
  • Ensure each subtask is self-describing; enough context exists for the Lead to create a TEP excerpt without re-examining the repo.
  • Default to the sectioned-v1 output format to facilitate safe excerpting and orchestration.

Quick Start

Provide a Task Packet to generate an ordered Delivery Plan with dependencies and a Definition of Done.

Frequently Asked Questions about delivery-plan

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

FAQPage Schema
How do I convert a task packet into an ordered delivery plan with dependencies?

To convert a task packet into a delivery plan, the Skill splits multi-domain work into single-domain subtasks, sequences them with dependencies and blockers, and generates a Definition of Done. This ensures clear handoffs and verifiable execution steps.

What is a Definition of Done in delivery planning and when do I need it?

A Definition of Done in delivery planning specifies the verifiable criteria for subtask completion. You need it when orchestrating execution across multi-domain tasks to ensure clear sequencing, handoffs, and self-describing steps for Leads.

How do I split multi-domain work into self-describing subtasks for execution?

Splitting multi-domain work into self-describing subtasks involves breaking down a task packet so each subtask contains enough context for a Lead to execute without re-examining the repository, ensuring safe excerpting and clear orchestration.

Does the sectioned-v1 output format support dependency and blocker tracking for subtasks?

Yes, the sectioned-v1 output format supports dependency and blocker tracking. It structures the ordered delivery plan by wrapping subtasks and metadata into a prescribed format that facilitates safe excerpting and orchestration.

Can I use this delivery planning approach for multi-domain tasks without external dependencies?

Yes, you can use this delivery planning approach for multi-domain tasks without external dependencies. It operates independently to convert task packets into ordered subtasks with dependencies and a Definition of Done.

Why does delivery planning require sequencing subtasks with blockers and dependencies?

Delivery planning requires sequencing subtasks with blockers and dependencies to establish clear execution order and handoffs. This allows Delivery PMs to orchestrate work safely and ensure each step is verifiable against the Definition of Done.