Spec-Driven Development
Also written: Specification-driven development, SDD
Software development guided by detailed specifications, often using AI to generate or verify code.
Spec-driven development means writing software based on very clear, detailed instructions. Think of it like building with a precise blueprint. Instead of just saying 'build a house,' you'd have exact measurements for every wall, window, and door.
In AI, this can mean using AI tools to help create that blueprint or even to write parts of the code based on the specifications. For example, a developer might describe a new feature in detail, and an AI could help generate the code for it, ensuring it matches the description.
Spec-driven development is a software engineering approach where the design and implementation of software are dictated by formal specifications. These specifications can range from high-level requirements to detailed interface definitions and logic. AI can be integrated into this process by generating code from specifications, verifying code against specifications, or even assisting in the creation of the specifications themselves.
This methodology aims to improve software quality, reduce ambiguity, and streamline development by providing a clear, verifiable target. Tradeoffs include the initial effort required to create comprehensive specifications and the potential for specifications to become outdated if not maintained rigorously.
Spec-driven development is gaining traction as AI models become more complex and integrated into enterprise systems. Companies are looking for structured ways to ensure AI outputs meet specific business requirements and quality standards. This approach helps manage the development lifecycle for AI applications, moving beyond simple prompts to robust engineering.
- From Spec-First to Enterprise-Ready: Extending GitHub Spec KitMicrosoft · Microsoft for Developers ·
- Expanding our partnership with CognizantAnthropic · Anthropic News ·
- Anthropic Claude
- Anthropic's partnership with Cognizant suggests an interest in embedding Claude into enterprise platforms, which likely involves specification-driven approaches for custom solutions.
- Microsoft Copilot
- Microsoft's GitHub Spec Kit is explicitly extending to support enterprise-level AI development, indicating a focus on specification-driven workflows for large organizations.
- Test-Driven Development
- TDD focuses on writing tests before code, whereas spec-
- Agile Development
- Agile emphasizes iterative development and flexibility, which
The concept of spec-driven development has roots in traditional software engineering methodologies like Model-Driven Architecture (MDA) and formal methods, which emphasize the use of detailed models and specifications to guide software creation. The integration of AI is a more recent evolution.
Ask an AI assistant like ChatGPT, Claude, or Gemini to write a simple Python function based on a detailed description. For example, 'Write a Python function that takes a list of numbers and returns the average, handling empty lists by returning 0.' Then, review the generated code to see if it precisely matches your specification.
The Playbook: reusable how-tos, prompts, configs and tips pulled out of the news.
- PromptRole-play a student struggling with math for AI tutor simulationSimulate a student's math struggles and uncertainty for AI tutor evaluation. · 5 min
- TemplateReusable workflow templates for healthcare teamsTeams can create and share templates for common workflows like documentation or prior authorizations. · 15 min
- TemplateTemplate for building an AI tool tailored to your workA structured approach to defining and building a custom AI tool for specific business needs. · 15 min
- TemplateEvals CSV prompt list for agent skillsA CSV file template for listing prompts used in evaluating agent skills. · 20 min
- HeuristicDouble coding agent task completion with docs linksProviding coding agents with a documentation link significantly increases their task completion rate. · 2 min
Questions people ask
- How does AI help with spec-driven development?
- AI can help by generating code directly from specifications, verifying that existing code meets specifications, or even assisting in writing more detailed and precise specifications.
- Is spec-driven development the same as just giving AI instructions?
- It's similar, but spec-driven development implies a more formal and detailed set of requirements or a blueprint that the AI adheres to, often within a structured development process.
First seen on Subvolts Jul 27, 2026. Updated Oct 14, 2026. Written by an AI model from the items above and reviewed against them weekly; report an error.