Table of Contents

AI Compatibility

Reqvire is built to be AI-friendly from the ground up.

Example AI Use Cases

1. Requirement Analysis

AI tools can analyze Markdown-based requirements to identify missing elements, suggest clearer phrasing, propose edge cases, and generate acceptance criteria or verification steps.

2. Architecture Suggestion

By understanding relationships in architecture files, AI can detect inconsistent component relationships, propose modular design improvements, and suggest missing interfaces or dependencies.

3. Traceability & Impact Prediction

AI tools can automatically trace which tests or requirements are impacted by changes, flag affected downstream areas, and summarize model diffs for engineering teams.

4. Test Coverage Assistance

Based on requirements and use cases, AI can recommend missing test scenarios and create verifications based on test criteria extracted from requirements chain.

5. Code Generation Assistance

AI can leverage structured requirements for:

  • Context-Aware Generation: Understanding the “why” behind every component through Reqvire’s traceability structure
  • Specification-Driven Coding: Generating code directly tied to specific requirements, reducing guesswork or misinterpretation
  • Verification-Backed Validation: Referring to defined verifications to ensure implementation meets intended behavior
  • Trace-Based Refactoring: Assisting in propagating requirement changes through the codebase and related artifacts

6. Report Generation

AI can automatically generate traceability reports, summarize system architecture, and prepare documentation for releases or reviews.

Human in Control — Always

While Reqvire empowers AI tools to act as smart collaborators, the human engineer remains the system’s captain — setting direction, making decisions, and approving outcomes.

Reqvire ensures that:

  • Every AI-suggested change is traceable and reviewable
  • System evolution remains understandable and documented
  • AI tooling acts in service of the engineer’s intent, not in place of it

Integration with AI Tools

Claude Code Plugin

Reqvire offers a dedicated Claude Code plugin that brings AI-assisted requirements engineering directly into your workflow with specialized skills and commands for model management.

AI Assistants and Copilots

Your Reqvire requirements work seamlessly with:

  • ChatGPT - Analyze requirements, suggest improvements, generate test scenarios
  • Claude - Deep requirement analysis, traceability review, impact assessment
  • GitHub Copilot - Context-aware code generation based on requirements
  • Cursor - Intelligent refactoring with requirement awareness
  • Local models - Local deployments can consume Reqvire models directly

Best Practices for AI-Assisted Requirements Engineering

1. Use Requirements as Context

Include relevant requirements files in your AI context to get specification-aware responses:

Analyze this code against requirements in specifications/Authentication.md

2. Leverage Traceability

Ask AI to follow requirement chains:

Show me all verifications for requirement REQ-AUTH-001 and its derived requirements

3. Generate Implementation Plans

Let AI create task breakdowns from requirements:

Create an implementation plan for the requirements in this file

4. Review Change Impact

Use AI to analyze how changes affect related elements:

If I modify requirement REQ-DATA-005, what else might be impacted?

5. Validate Consistency

Ask AI to check for inconsistencies:

Review these requirements for conflicts or missing relations

Why Reqvire Works Well with AI

Structured but Readable

  • Semi-structured Markdown is easily parsed by AI tools
  • No complex DSLs or proprietary formats
  • Natural language requirements with clear structure

Explicit Relationships

  • Clear parent-child hierarchies
  • Defined verification links
  • Traceable implementation connections

Self-Documenting

  • Requirements serve as both spec and documentation
  • Relations make dependencies explicit
  • History preserved in Git

Context-Rich

  • Each element contains description and rationale
  • Relations provide full context graph
  • Verifications define expected behavior