AI-DevOps

Unified approach to AI-assisted software development and DevOps, aggregating research and best practices from multiple projects.

Purpose

This project consolidates instruction files, prompts, and workflows for AI-assisted development from several mature projects:

• minouris/spafw37 (most comprehensive workflow)
• minouris/prompt-driven-development (metaprompts and composition)
• minouris/claude-code-container (most up-to-date instructions)
• minouris/nightingale-truenas (memory-based approach)

Quick Start

1. Read the Analysis: analysis.md contains a comprehensive analysis of all source projects with timeline, maturity assessment, and key findings.

2. Review Recommendations: recommendations.md provides the concrete import plan with 6 phases and specific files to copy.

3. Follow the Import Plan: Start with Phase 1 (core meta-instructions) and work through successive phases.

Status

✅ Completed: Research and analysis phase (Issue #1)
🔲 Next: Begin Phase 1 imports (core instructions)

Documentation

• analysis.md - Full project analysis, timeline, and maturity assessment
• recommendations.md - Concrete import plan and implementation guide

Structure (Planned)

ai-devops/
├── .github/
│   ├── copilot-instructions.md
│   ├── instructions/          # How to do things
│   │   ├── core/              # Meta-instructions
│   │   ├── standards/         # Quality standards
│   │   ├── technical/         # Domain-specific
│   │   └── composition/       # Advanced patterns
│   ├── prompts/               # What to do
│   │   ├── planning/
│   │   └── execution/
│   └── agents/                # Specialized AI personas
├── docs/
│   ├── framework/             # System documentation
│   ├── lessons/               # Field lessons
│   └── research/              # Background research
└── README.md

Key Principles

1. Selective Loading: Only load instructions relevant to current task
2. Step-Based Execution: Break plans into discrete, trackable steps
3. Memory Management: Explicit fact tracking and verification
4. Continuous Refinement: Field lessons fed back into instructions
5. Safety First: Built-in verification and guardrails
6. Tool-Agnostic Core: But with tool-specific optimizations available

Contributing

This is a consolidation project. Improvements should generally be:

1. Tested in this project
2. Fed back to source projects as appropriate
3. Documented as field lessons