A modern developer playbook for GitHub, AI-assisted development, and production-ready workflows.
See all workflows as diagrams: docs/00-start-here/VISUAL_GUIDE.md
Follow the numbered steps. By the end you will be working like a professional developer.
Skip the basics:
| Goal | Go To |
|---|---|
| Daily workflow + PR process | docs/03-development-workflow/ |
| AI tool integration | docs/04-ai-workflows/ |
| AI operating layer for agents, skills, prompts, context, and governance | ai/ |
| Tooling ecosystem, MCP, and external references | docs/12-tooling-ecosystem/ |
| Team standards and enforcement | docs/06-standards/ |
| GitHub Actions enforcement pack | docs/10-github-actions/GITHUB_ACTIONS_ENFORCEMENT_PACK.md |
| Compliance (NIST/FIPS) | docs/07-compliance/ |
| Verified security mistakes, PoCs, and the checklist they earned | docs/13-lessons-learned/LESSONS_LEARNED.md |
Copy templates into your repository:
templates/project/ ← core project docs (DESIGN, TESTING, DEPLOYMENT, SECURITY)
templates/workflows/ ← GitHub Actions starters
templates/prompts/ ← AI prompt templates
templates/tools/ ← external repo, tool, and LLM review templates
templates/mcp/ ← MCP server and MCP security review templates
See complete working examples:
- GitHub from beginner to professional workflows
- Branch → Commit → Push → Pull Request → Merge — with real examples
- Development environment setup (Windows, Mac, VS Code, Codespaces)
- AI-assisted development with human-in-the-loop guardrails
- Tool and MCP intake using a reference-first review model
- CI/CD enforcement via GitHub Actions
- Enterprise and government-ready practices (NIST/FIPS)
Most repos give you code. This gives you:
- Structured workflows — the professional development cycle documented end to end
- Reusable templates — copy-paste starting points that work immediately
- AI-safe practices — AI generates, humans review and decide — enforced, not just suggested
- AI operating layer — agents, skills, prompts, context, and governance are separated from human learning docs
- Tool intake guardrails — external repos, tools, MCP servers, and models are linked first and reviewed before adoption
- Automated enforcement — GitHub Actions checks that catch problems before merge
- Living updates — a system that adapts as GitHub, Anthropic, OpenAI, and Google ship changes
This repository uses GitHub Actions to enforce standards automatically on every pull request:
| Workflow | What It Enforces |
|---|---|
00-repo-health.yml |
Required files present (SECURITY.md, AGENTS.md, CODEOWNERS) |
01-pr-standards.yml |
PR description filled out; no WIP merges |
02-docs-quality.yml |
No broken links; markdown lint passes |
03-security-supply-chain.yml |
No secrets committed; dependency audit |
04-codeql.yml |
Static code analysis for security vulnerabilities |
07-branch-naming.yml |
Branch name follows type/description convention |
08-commit-lint.yml |
All commits follow Conventional Commits format |
Bypass is not the answer. When a check fails, fix the underlying issue — do not use --no-verify or skip the workflow.
→ Full details: docs/10-github-actions/GITHUB_ACTIONS_ENFORCEMENT_PACK.md
→ Full directory listing: REPO_MAP.md
docs/ Human-facing documentation: getting started, workflows, AI tools, compliance
ai/ AI-facing operating layer: agents, skills, prompts, context, governance
templates/ Copy-paste starting points for projects, workflows, prompts, MCP, and tool review
examples/ Working examples showing templates in real project structures
scripts/ Repo validation and maintenance scripts
.github/ GitHub Actions workflows, issue templates, PR template
Key root files:
| File | Purpose |
|---|---|
AGENTS.md |
Rules and boundaries for AI coding agents |
TECH_STACK.md |
Technology choices and rationale |
DESIGN.md |
Architecture and design decisions |
SECURITY.md |
Security policies and responsible disclosure |
Detected by the Living Updates system — updated weekly. Full notes in docs/06-living-updates/incoming/.
| Date | Source | Update |
|---|---|---|
| 2026-08-10 | Anthropic | Claude Sonnet 5 and Opus 5 now GA — current recommended models for coding |
| 2026-08-10 | Anthropic | Managed Agents GA — server-hosted agents with persistent memory and skill composition |
| 2026-08-10 | Anthropic | Files API GA — upload documents once, reference by ID across requests |
| 2026-08-10 | Anthropic | Extended + interleaved thinking GA — model reasoning between tool calls |
| 2026-08-10 | Anthropic | Tool Runner GA — SDK handles tool-use loop automatically |
| 2026-08-10 | Anthropic | MCP (Model Context Protocol) open standard — growing ecosystem of server integrations |
| 2026-05-03 | GitHub Copilot | GPT-5.5 now GA — strongest on multi-step agentic coding tasks |
| 2026-05-03 | GitHub Copilot | Copilot code review will consume Actions minutes from June 1 |
| 2026-05-03 | GitHub Copilot | Inline agent mode in preview for JetBrains IDEs |
| 2026-05-03 | OpenAI | API changelog updated — check for model and rate limit changes |
→ Review all incoming notes · How the system works · Tracked sources
Top vendor-certified paths for AI and ML. Full details, costs, and status tracking in docs/04-ai-workflows/AI_CERTIFICATIONS.md.
| Cert | Vendor | Level | Cost | Status | Replaced By |
|---|---|---|---|---|---|
| Azure AI Fundamentals (AI-900) | Microsoft | Beginner | ~$165 | 🔄 Retiring Jun 2026 | AI-901 |
| Azure AI Fundamentals (AI-901) | Microsoft | Beginner | ~$165 | 🆕 Launching Jun 2026 | — |
| Azure AI Engineer Associate (AI-102) | Microsoft | Associate | ~$165 | 🔄 Retiring Jun 2026 | AI-103 |
| Azure AI Services Agent Associate (AI-103) | Microsoft | Associate | ~$165 | 🆕 Launching Jun 2026 | — |
| Applied Skills — GenAI / Prompt Engineering | Microsoft | Task-based | Free | ✅ Active | — |
| AWS Certified AI Practitioner (AIF-C01) | AWS | Foundational | ~$100 | ✅ Active | — |
| AWS ML Engineer Associate (MLA-C01) | AWS | Associate | ~$150 | ✅ Active | — |
| AWS ML Specialty (MLS-C01) | AWS | Specialty | ~$300 | 🔄 Retired Mar 2026 | AIP-C01 |
| AWS AI Practitioner+ (AIP-C01) | AWS | Associate | ~$150 | 🆕 2026 | — |
| Generative AI Leader | Google Cloud | Professional | ~$99 | 🆕 May 2025 | — |
| Professional ML Engineer | Google Cloud | Professional | ~$200 | ✅ Active | — |
| OCI Generative AI Professional | Oracle | Professional | ~$245 | ✅ Active | — |
| CompTIA SecAI+ (CY0-001) | CompTIA | Intermediate | ~$239 | 🆕 Feb 2026 | — |
| DLI — Deep Learning / GenAI | NVIDIA | Course | ~$30–90 | ✅ Active | — |
Free learning (no cert): fast.ai · Hugging Face · DeepLearning.AI · Google ML Crash Course
→ Choosing a path + full cert list
A developer who works through this repository will be able to:
- Set up a full development environment from scratch
- Work confidently with Git and GitHub
- Build and manage real projects using professional workflows
- Use AI tools responsibly and effectively
- Apply enterprise-grade standards without rewriting them from scratch
Reusable documentation and workflow scaffold for AI-assisted software development.
Add setup and run steps for this repository.
Active development.