diff --git a/README.md b/README.md
index 911555e..a089c03 100644
--- a/README.md
+++ b/README.md
@@ -1,78 +1,113 @@
# Ninobyte
-> Enterprise tooling for Claude agent ecosystems.
+> Governance-first agent tooling, AWS-native training labs, and applied AI infrastructure.
[](https://github.com/iamnortey/portfolio/blob/main/case-studies/ninobyte.md)
[](https://github.com/iamnortey/portfolio)
---
-## What It Does
+## What it is
-Ninobyte provides enterprise-grade education and tooling for Anthropic/Claude agent ecosystems. Verified, secure, and reproducible skill packs, MCP servers, and Claude Code plugins — with full validation trails.
+Ninobyte is a multi-product effort to bring engineering rigor to applied AI work. It spans three connected product lines:
-**Validation-first approach:** Every skill validated against official Anthropic documentation. Every decision traced to evidence.
+1. **Agent tooling** — Claude Skills, MCP servers, and Claude Code plugins with full validation trails against official Anthropic documentation.
+2. **AWS-native training labs** — governed, evidence-based AI CloudOps and security training, delivered through ticket-driven job simulation and proof-pack-style portfolio artifacts.
+3. **Applied AI infrastructure** — schemas, governance, and product systems for AI-native applications in education and regulated contexts.
+
+The throughline across all three is the same: evidence-based engineering, governance before deployment, and proof of work over passive certificates.
---
-## The Problem
+## The problem
+
+- "Applied AI" content is largely unverified — skills, MCP servers, and tutorials drift from official documentation
+- Cloud AI courses teach vocabulary without giving learners a safe place to operate workloads
+- Education data work in emerging markets often skips rights, schema, and governance discipline
+- Practitioners and reviewers need auditable evidence, not asserted expertise
-- Official documentation is sparse for advanced use cases
-- Most tutorials are unverified and may be outdated
-- Enterprise organizations need auditable agent tooling
-- Skills, MCP, and Claude Code have different patterns that are easy to conflate
+Ninobyte addresses these by treating governance, validation, and evidence as first-class deliverables across every product line.
---
-## Products
+## Product lines
-| Product | Description |
-|---------|-------------|
-| **Senior Developer's Brain** | Job system for enterprise engineering workflows |
-| **MCP Server Templates** | Boilerplate for MCP server development |
-| **Claude Code Plugins** | Extensions for Claude Code |
+### Agent tooling
----
+Lives in the private parent repository [`iamnortey/ninobyte`](https://github.com/iamnortey/ninobyte) (private).
-## Stack
+| Component | Purpose |
+|---|---|
+| **Senior Developer's Brain** | Job system for enterprise engineering workflows — also available as a Claude Code plugin |
+| **MCP Server Templates** | Boilerplate and patterns for MCP server development |
+| **Claude Code Plugins** | Extensions packaged for Claude Code |
+| **Skill packs** | Validated against official Anthropic documentation before release |
-| Component | Technology |
-|-----------|------------|
-| **Skills** | Claude Skills format |
-| **MCP** | Model Context Protocol |
-| **Plugins** | Claude Code extensions |
-| **Language** | Python |
+### AWS-native training labs
----
+Lives in the public organization [`ninobyte-cloudops-lab`](https://github.com/ninobyte-cloudops-lab).
-## Key Patterns
+| Lab | Audience | Status |
+|---|---|---|
+| [**AI-Native CloudOps Lab**](https://github.com/ninobyte-cloudops-lab/cloudops-lab-overview) | Cloud builders, operators, emerging AWS/AI engineers, defensive security learners | Platform foundation; live AWS cohort readiness gated |
+| [**AI Security & Governance Lab — AWS Edition**](https://github.com/ninobyte-cloudops-lab/ai-security-governance-lab-overview) | Cloud security engineers, GRC analysts, IT auditors, security managers, AI governance pros | Docs-first foundation complete; AWS execution gated |
+| [**Student workspace model**](https://github.com/ninobyte-cloudops-lab/student-workspace-preview) | Enrolled learners and cohort participants | Private template beta-ready; delivery workflow pending |
-### Validation-First
-Every skill validated against official Anthropic sources before release.
+### Applied AI infrastructure
-### Governance Versioning
-Semantic versioning for governance documents, not just code.
+Lives in the public-profile organization [`ninobyte-labs`](https://github.com/ninobyte-labs) (org profile public; product repos private).
-### Evidence Trails
-All decisions link back to canonical documentation.
+| Project | Purpose |
+|---|---|
+| **Ghana Education Data OS (GEDOS)** | Governed control plane for Ghana curriculum and exam intelligence — JSON Schemas, governance and rights policy, review/promotion lifecycle |
+| **GEDOS Teacher Portal** | Fumadocs + Payload teacher resource portal |
+| **Teacher-to-Author Lab** *(concept stage)* | Training to help Ghanaian teachers turn governed exam intelligence into original lesson notes |
+| **Applied AI product engineering** | Schemas and product systems for AI-native applications in regulated contexts (e.g., the metabolic-platform mobile guidance project) |
---
-## Documentation
+## Key patterns
-| Document | Description |
-|----------|-------------|
-| [Case Study](https://github.com/iamnortey/portfolio/blob/main/case-studies/ninobyte.md) | Full project overview |
+| Pattern | What it means in practice |
+|---|---|
+| **Validation-first** | Every skill is validated against official Anthropic sources before release. Every claim links back to documentation. |
+| **Governance versioning** | Governance documents get semantic versioning, not just code. Policies are auditable changes, not unwritten norms. |
+| **Evidence trails** | Decisions are recorded in ADRs and validation logs. Lab work produces sanitized proof packs. |
+| **Phase-gated development** | No Phase 2 until Phase 1 passes all quality gates. Applies to product, labs, and data work alike. |
+| **Public/private boundary discipline** | Public-track repos hold schemas, documentation, diagrams, and safe samples. Implementation details, learner materials, instructor keys, and rights-sensitive data stay private. |
---
-## Access
+## Ecosystem at a glance
+
+```mermaid
+flowchart LR
+ A[Ninobyte] --> B[Agent tooling]
+ A --> C[AWS-native training labs]
+ A --> D[Applied AI infrastructure]
+
+ B --> B1[iamnortey/ninobyte
private]
+ C --> C1[ninobyte-cloudops-lab
public org]
+ D --> D1[ninobyte-labs
public org]
+```
+
+---
+
+## What's public vs private
+
+This repository contains architecture documentation, governance patterns, ADR examples, and validation methodology at a strategic level. Implementation details are split across three repositories:
+
+- [`iamnortey/ninobyte`](https://github.com/iamnortey/ninobyte) — private parent for agent tooling (Senior Developer's Brain, MCP servers, Claude Code plugins)
+- [`ninobyte-cloudops-lab`](https://github.com/ninobyte-cloudops-lab) — public org with three public overview repositories; product repos remain private
+- [`ninobyte-labs`](https://github.com/ninobyte-labs) — public org profile; product repos (GEDOS, teacher portal, metabolic-platform) remain private
-The core implementation is in a **private repository**. This repository contains architecture documentation, governance patterns, ADR examples, and validation methodology.
+For partnership, cohort, or technical discussions, reach Ninobyte through its official channels.
---
## Related
-- [Portfolio](https://github.com/iamnortey/portfolio) — all case studies and architecture samples
+- [Portfolio](https://github.com/iamnortey/portfolio) — case studies, architecture, ADRs, runbooks
- [Case Study](https://github.com/iamnortey/portfolio/blob/main/case-studies/ninobyte.md) — full project deep-dive
+- [`ninobyte-cloudops-lab`](https://github.com/ninobyte-cloudops-lab) — AWS training labs org
+- [`ninobyte-labs`](https://github.com/ninobyte-labs) — applied AI and education data org