aif-plan

Generate structured feature implementation plans with tasks, branches, and milestones.

Updated Mar 22, 2026
One-click install
npx skills add https://github.com/afterwork/arra-aif --skill aif-plan-afterwork
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: aif-plan
Source: https://github.com/afterwork/arra-aif/tree/main/.cursor/skills/aif-plan
Command: npx skills add https://github.com/afterwork/arra-aif --skill aif-plan-afterwork

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Plan and coordinate implementation tasks for features, aligning branches, roadmaps, and project context to enable smooth delivery.

Core Features & Use Cases

  • Generates structured implementation plans for features and tasks, including Fast and Full modes with optional parallel worktrees.
  • Integrates project context from ROADMAP.md and RESEARCH.md to tailor plans, milestones, and decisions.
  • Produces plan files in .ai-factory/PLAN.md (fast) or .ai-factory/plans/<branch>.md (full) with explicit tasks, checkpoints, and logging requirements.

Quick Start

Ask /aif-plan to generate a plan for a new feature and choose between full or fast modes.

Frequently Asked Questions about aif-plan

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

FAQPage Schema
How do I plan feature implementation tasks with branch-aware workflows?

Feature implementation planning with branch-aware workflows is generated by outlining tasks, branches, and milestones from a given project context. You choose between full planning with parallel worktrees or fast planning without branches.

How does feature planning integrate with project roadmaps and research context?

Feature planning integrates with project roadmaps and research context by reading ROADMAP.md and RESEARCH.md files. It tailors plans, milestones, and decisions based on the context found in those files.

What is the difference between fast and full modes for generating plan files?

Fast mode generates plan files in .ai-factory/PLAN.md without branches, while full mode creates detailed plans in .ai-factory/plans/<branch>.md with explicit tasks, checkpoints, and optional parallel worktrees.

Can I use parallel worktrees for task management in feature planning?

Yes, you can use parallel worktrees for task management by selecting full mode during feature planning. This branch-aware approach structures tasks and milestones to enable parallel delivery workflows.

Do I need specific project context files to generate a structured implementation plan?

While feature planning can run without them, having ROADMAP.md and RESEARCH.md files in your project context allows the planner to tailor milestones and decisions to your specific roadmap and research findings.

What are the limitations of fast planning mode without branches?

Fast planning mode skips branch-aware parallel worktrees, generating a single .ai-factory/PLAN.md file. It is suited for rapid task outlining but lacks the detailed checkpoints and branch-specific milestones of full mode.