Skip to content

markhazlett/cursor-rules

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

2 Commits
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

cursor-rules

A set of cursor rules I'm using and experimenting with

Overview

This repository contains a collection of Cursor AI rules designed to streamline software development workflows. These rules guide AI assistants through structured processes for creating Product Requirements Documents (PRDs) and generating detailed task lists for implementation.

Rules Collection

πŸ“‹ create_prd.mdc

Purpose: Generate comprehensive Product Requirements Documents from initial user prompts

What it does:

  • Guides AI through asking clarifying questions before writing PRDs
  • Creates structured PRDs with clear sections (goals, user stories, functional requirements, etc.)
  • Targets junior developers as the primary audience
  • Saves PRDs to /tasks/prd-[feature-name].md

Key Features:

  • Interactive clarification process
  • Standardized PRD structure with 9 key sections
  • Emphasis on clear, actionable requirements
  • Scope management with explicit non-goals

πŸ“ process_task_list.mdc

Purpose: Convert PRDs into detailed, step-by-step implementation task lists

What it does:

  • Analyzes existing PRDs to extract implementation requirements
  • Creates hierarchical task lists with parent tasks and sub-tasks
  • Identifies relevant files that need to be created or modified
  • Uses a two-phase approach (high-level tasks β†’ detailed sub-tasks)

Key Features:

  • Interactive confirmation workflow ("Go" to proceed)
  • Structured output format with checkboxes
  • File identification with test file recommendations
  • Junior developer-focused instructions

🎯 generate_tasks_tracer.mdc

Purpose: Advanced task generation using the "tracer bullet" methodology

What it does:

  • Implements tracer bullet development approach
  • Creates minimal end-to-end feature slices for rapid validation
  • Separates tracer bullet tasks (TB-) from production readiness tasks (PR-)
  • Multi-phase workflow with validation checkpoints

Key Features:

  • Tracer bullet methodology for rapid prototyping
  • Two-phase development: minimal viable slice β†’ production hardening
  • Interactive gates with specific commands ("Go TB", "Go PR", "Go PR Sub")
  • Focus on visible, working features before polish

πŸ“„ generate_tasks.mdc

Purpose: Simplified version of task list generation

What it does:

  • Streamlined task generation from PRDs
  • Similar to process_task_list.mdc but with a more concise approach
  • Focuses on core task breakdown without advanced methodologies

Usage

These rules are designed to work with Cursor AI to:

  1. Start with a feature idea β†’ Use create_prd.mdc to create a comprehensive PRD
  2. Convert PRD to tasks β†’ Use either:
    • process_task_list.mdc for standard task breakdown
    • generate_tasks_tracer.mdc for tracer bullet methodology
    • generate_tasks.mdc for simplified task generation

File Structure

All generated documents are saved to /tasks/ directory:

  • PRDs: prd-[feature-name].md
  • Task lists: tasks-[prd-file-name].md

Target Audience

These rules assume the implementation will be done by junior developers, so they emphasize:

  • Clear, unambiguous instructions
  • Detailed step-by-step breakdowns
  • Explicit file identification and testing guidelines
  • Structured, consistent output formats

About

A set of cursor rules I'm using and experimenting with

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published