AirSDLC¶
The AI-Responsible Software Development Lifecycle
A comprehensive framework specification for AI-driven software development that bridges the gap between business requirements and production code.
Quick Navigation¶
-
π Get Started
New to AirSDLC? Start here to understand the framework.
-
π§ Core Philosophy
Learn the foundational AI-DLC principles behind AirSDLC.
-
π Three Phases
Understand Inception, Design, and Construction phases.
-
π Artifacts
Complete specifications for DAA, ADR, TIP, RFC, and more.
-
β Workflow Guide
Step-by-step process from PRD to production.
-
β‘ Operations
Context-aware incident response and monitoring.
-
π§© Extensibility
Adapt the framework to your team's unique needs.
-
π» Examples
See real-world artifacts in action.
What is AirSDLC?¶
AirSDLC (AI-Responsible Software Development Lifecycle) is not a toolβit is a framework specification that provides:
- Core Principles: AI-driven execution with human oversight
- Artifact Definitions: Structured specifications for all lifecycle artifacts
- Lifecycle Phases: Sequential knowledge handoff from business to code
- Workflow Patterns: Repeatable, validated processes
- Extension Guidelines: Adapt to your team's needs
The Three Phases¶
graph LR
A[Business Intent] --> B(Inception)
B --> C[AI-DAA]
C --> D(Design)
D --> E[AI-ADR]
E --> F(Construction)
F --> G[Production Code]
- Inception (The "WHAT") - Business intent β Domain model
- Design (The "HOW") - Domain model β Architecture decisions
- Construction (The "BUILD & RUN") - Decisions β Production code
Framework vs. Implementation¶
Important Distinction
- AirSDLC (this repository): The framework specification
- Implementation Tools: Concrete software that implements the framework
This separation ensures multiple tools can implement the same standard.
Framework Origins
AirSDLC is an open-source implementation of the AI-Driven Development Lifecycle (AI-DLC) framework developed by Amazon Web Services. See AI-DLC Attribution for details.
For Different Audiences¶
Getting Started:
- Read Overview for the big picture
- Study Philosophy for core principles
- Follow Workflow for practical guidance
- Explore Examples to see artifacts
Best Practices:
- Start with Lightweight Workflow for simple features
- Use Full Workflow for complex/high-risk features
- Maintain Knowledge Repository as single source of truth
- Validate AI outputs at every phase
Implementation Guide:
- Review Artifacts for data structures
- Study Lifecycle for phase requirements
- Check Extensibility for conformance criteria
- Build tools that implement this specification
Conformance Requirements:
- β Implement core artifacts (PRD, DAA/TIP, ADR)
- β Support sequential phases with validation gates
- β Maintain traceability chain
- β Store artifacts in Knowledge Repository
Strategic Adoption:
- Evaluate Philosophy against team values
- Customize Extensibility for your context
- Build Architectural Playbook
- Define workflow paths for your complexity matrix
Key Decisions:
- Full vs Lightweight workflow thresholds
- Custom artifact types for compliance
- Validation ceremony cadence
- Tool selection criteria