grill-plan

Review autosprint/plan.md tasks against a rubric for agent readiness.

5|4|Updated Apr 8, 2026
One-click install
npx skills add https://github.com/haakonbull/autosprint --skill grill-plan-haakonbull
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: grill-plan
Source: https://github.com/haakonbull/autosprint/tree/main/.claude/skills/grill-plan
Command: npx skills add https://github.com/haakonbull/autosprint --skill grill-plan-haakonbull

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill ensures that your autosprint plan is meticulously pre-reviewed for clarity and readiness, eliminating ambiguities and improving efficiency.

Core Features & Use Cases

  • Task Review: Examines each task in your plan.md for completeness and execution readiness.
  • Quality Assurance: Flags and refines tasks that are unclear, overly complex, or not properly scoped.
  • Use Case: Before executing a new autosprint, use grill-plan to review and refine tasks, ensuring they can be accurately and effectively implemented by an autosprint agent.

Quick Start

Run 'grill-plan' to automate the review and enhancement of tasks in your autosprint/plan.md.

Frequently Asked Questions about grill-plan

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

FAQPage Schema
How do I review my autosprint plan to ensure tasks are ready for an AI agent?▼

Review your autosprint plan by applying a rubric of requirements like concreteness, scoping, and done-conditions. This ensures tasks in your plan.md are clear, self-contained, and agent-ready before execution.

What makes a task in a software development workflow agent-ready?▼

An agent-ready task in a software development workflow must be concrete, properly scoped, have clear done-conditions, be self-contained, and align with the destination. Distinctness from other tasks is also required.

How do I refine unclear tasks in my plan.md without re-planning the entire project?▼

Refine unclear tasks in your plan.md by operating on existing entries without re-planning. Flag and refine tasks that are overly complex or not properly scoped to improve execution readiness.

Can I check if my autosprint tasks meet specific quality conditions before running them?▼

Yes, you can check if your autosprint tasks meet quality conditions by examining them for completeness. A rubric verifies realness, destination-alignment, and distinctness to ensure they can be accurately implemented.

What are the limitations of using AI-assisted planning for task optimization?▼

AI-assisted planning for task optimization operates on existing tasks without re-planning. It is intended strictly for preparing autosprint runs and will not generate a new plan if tasks are fundamentally misaligned.

Do I need any specific dependencies to prepare my plan for an autosprint run?▼

No specific dependencies are required to prepare your plan for an autosprint run. You only need an existing autosprint plan.md file to run the review and enhancement scripts on your tasks.