Fix README conflict markers

README.md

@@ -11,7 +11,59 @@ website/          Docusaurus frontend (learn.flowresearch.tech)
 scripts/          AI-assisted content generation tools
 ```
 
-## Quick start
+---
+
+## 🤖 AI Agent Instructions (System Context)
+
+**If you are an AI Agent interacting with this repository, adhere to these constraints:**
+
+### 1. The Three Technical Pillars
+Every piece of content must align with one or more of these pillars:
+*   **AI/ML:** Federated Learning (Flower, PySyft), Decentralized Training, Model Privacy.
+*   **Blockchain:** Infrastructure (Ethereum, Filecoin, Bittensor), Incentive Design.
+*   **Protocol Engineering:** P2P Networking (libp2p), Distributed Storage (IPFS), Compute (Akash/Gensyn).
+
+### 2. The Learning Hierarchy
+Our curriculum uses precise track-tier-section-lesson organization for clarity.
+
+- `01-foundations/`
+  - sections directly at root: `01-concepts`, `02-practice`, `03-tooling`
+  - lessons like `01-learning-hierarchy.md`, `02-effective-notes.md`, `03-collaboration-workflows.md`
+
+- `02-blockchain/`, `03-ai-ml/`, `04-protocol-eng/`
+  - tier folders: `beginner`, `intermediate`, `advanced`
+  - each tier has domain sections (e.g., blockchain advanced has `01-protocol-engineering`, `02-scalability`)
+  - each section contains lesson markdown files named by topic (e.g., `01-protocol-architecture.md`)
+
+- This structure is intended to map directly to competency progression:
+  - **beginner**: concepts and fundamentals
+  - **intermediate**: applied building and safety
+  - **advanced**: architecture, deployment, ecosystem leadership
+
+### 3. Voice & Tone
+*   **Audience:** Mid-level Software Engineers.
+*   **Tone:** Engineering-first, technical, concise. Avoid marketing fluff.
+*   **Context:** Use African-centric technical examples where possible (e.g., low-bandwidth optimization, local payment gateways).
+
+### 4. Content Automation Loop
+The `scripts/` directory is designed to:
+1.  Scrape external protocol documentation (e.g., Protocol Labs, Flower Labs).
+2.  Generate new lessons or blog articles based on documentation updates.
+3.  Rewrite outdated `labs/` code snippets when a protocol updates its API.
+4.  **Requirement:** All generated Markdown MUST include YAML frontmatter with `id`, `title`, `track`, `level`, and `version`.
+
+---
+
+## 🛠️ Tech Stack
+*   **Content:** Markdown (MDX)
+*   **Frontend:** Docusaurus (React-based SSG)
+*   **Automation:** Python (OpenAI/Anthropic APIs, BeautifulSoup for scraping)
+*   **Infrastructure:** GitHub Actions, GitHub Pages/Vercel
+
+---
+
+## 📜 Contribution & Policy
+All lessons and research articles go through a **Human-in-the-Loop** review process. AI-generated content is staged in a Pull Request for review by the technical team. 
 
 ```bash
 cd website