diff --git a/README.md b/README.md
index 9245672..54e2369 100644
--- a/README.md
+++ b/README.md
@@ -2,7 +2,7 @@
 
 ![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)
 
-A 100-line minimalist LLM framework for agents, task decomposition, RAG, etc.
+A [100-line](minillmflow/__init__.py) minimalist LLM framework for agents, task decomposition, RAG, etc.
 
 - Install via  ```pip install minillmflow```, or just copy the [source](minillmflow/__init__.py) (only 100 lines)
 
@@ -27,7 +27,7 @@ Hence, I built this framework that lets LLMs focus on what matters. It turns out
 
 
 <div align="center">
-  <img src="./assets/minillmflow.jpg" width="400"/>
+  <img src="/assets/minillmflow.jpg" width="400"/>
 </div>
 
 ## Example
diff --git a/docs/index.md b/docs/index.md
index 8bf1275..0a65aa4 100644
--- a/docs/index.md
+++ b/docs/index.md
@@ -1,23 +1,26 @@
 ---
 layout: default
 title: "Home"
+nav_order: 1
 ---
 
 # Mini LLM Flow
 
 A 100-line minimalist LLM framework for agents, task decomposition, RAG, etc.
 
-![Alt text](/docs/assets/minillmflow.jpg)
+<div align="center">
+  <img src="/assets/minillmflow.jpg" width="400"/>
+</div>
 
 ## Core Abstraction
 
 We model the LLM workflow as a **Nested Flow**:
-
-- Each **Node** handles a simple LLM task (e.g., text summarization, structure extraction, or question answering).
-- Nodes are chained together to form a **Flow** for more complex tasks. One Node can be chained to multiple Nodes based on **Actions**, e.g., for agentic steps.
+- Each **Node** handles a simple LLM task.
+- Nodes are chained together to form a **Flow** for more complex tasks.
+- One Node can be chained to multiple Nodes based on **Actions**.
 - A Flow can be treated as a Node for **Nested Flows**.
-- Both Nodes and Flows can be **Batched** for data-intensive tasks (e.g., processing one file at a time in a directory of files).
-- Nodes and Flows can be **Async**, e.g., for user feedback before proceeding.
+- Both Nodes and Flows can be **Batched** for data-intensive tasks.
+- Nodes and Flows can be **Async**.
 
 - [Node](./node.md)
 - [Flow](./flow.md)
diff --git a/docs/node.md b/docs/node.md
index 7552496..265eeac 100644
--- a/docs/node.md
+++ b/docs/node.md
@@ -1,6 +1,7 @@
 ---
 layout: default
 title: "Node"
+nav_order: 2
 ---
 
 # Node
@@ -38,16 +39,16 @@ When an exception occurs in `exec()`, the Node automatically retries until:
 
 If you want to **gracefully handle** the error rather than raising it, you can override:
 
-``  
+```
 def process_after_fail(self, shared, prep_res, exc):
     raise exc
-``  
+```
 
 By **default**, it just re-raises `exc`. But you can return a fallback result instead. That fallback result becomes the `exec_res` passed to `post()`.
 
 ## Minimal Example
 
-``  
+```
 class SummarizeFile(Node):
     def prep(self, shared):
         filename = self.params["filename"]
@@ -68,5 +69,5 @@ class SummarizeFile(Node):
         filename = self.params["filename"]
         shared["summary"][filename] = exec_res
         # Return "default" by not returning anything
-``  
+```