requirements-management

Manage structured markdown project requirements with stable IDs and status flow.

11|Updated Feb 8, 2026
One-click install
npx skills add https://github.com/njt/reqskill --skill requirements-management-njt
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: requirements-management
Source: https://github.com/njt/reqskill/tree/main/skills/requirements-management
Command: npx skills add https://github.com/njt/reqskill --skill requirements-management-njt

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill addresses the challenge of keeping project requirements accurate and synchronized with evolving code, preventing LLM drift and ensuring the system aligns with user needs.

Core Features & Use Cases

  • Structured Requirements: Defines a clear format for requirements.md including sections for problem statements, actors, glossary, functional requirements, constraints, decisions, and open questions.
  • Workflow Integration: Provides guidance for starting new projects, coding sessions, and updating requirements, emphasizing the confirm-delegate-implement pattern.
  • Use Case: When starting a new coding session, use this Skill to load and review the requirements.md file to ensure your implementation aligns with the project's defined capabilities and constraints.

Quick Start

Use the requirements-management skill to review the project's requirements.md file.

Frequently Asked Questions about requirements-management

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

FAQPage Schema
How do I manage project requirements for LLM-assisted development?

Manage project requirements by maintaining a structured `requirements.md` file that synchronizes constraints and decisions with evolving code, preventing LLM drift. It serves as a durable source of truth throughout the software development lifecycle.

What is the best way to structure a requirements document for AI coding sessions?

The best way to structure a requirements document is using a markdown format with specific sections for problem statements, actors, glossary, functional requirements, constraints, decisions, and open questions to ensure clear AI context.

How do I prevent LLM drift when updating functional requirements?

Prevent LLM drift by enforcing stable IDs, status flow, and an append-only decisions log within your requirements document. This maintains a durable source of truth that keeps implementations aligned with defined capabilities.

Can I use markdown for tracking software engineering decisions and open questions?

Yes, you can use markdown for tracking software engineering decisions and open questions. The format enforces an append-only decisions section to maintain a durable source of truth throughout the project lifecycle.

How do I start a new coding session using a structured requirements file?

Start a new coding session by loading and reviewing the `requirements.md` file. This ensures your implementation aligns with the project's defined capabilities and constraints before initiating the confirm-delegate-implement workflow pattern.