diff --git a/docs/async.md b/docs/async.md
index e366327..589234e 100644
--- a/docs/async.md
+++ b/docs/async.md
@@ -1,7 +1,8 @@
 ---
 layout: default
 title: "Async"
-nav_order: 7
+parent: Core Abstraction
+nav_order: 5
 ---
 
 # Async
diff --git a/docs/batch.md b/docs/batch.md
index 540de5a..bc53d00 100644
--- a/docs/batch.md
+++ b/docs/batch.md
@@ -1,7 +1,8 @@
 ---
 layout: default
 title: "Batch"
-nav_order: 6
+parent: Core Abstraction
+nav_order: 4
 ---
 
 # Batch
diff --git a/docs/communication.md b/docs/communication.md
index b4115f6..3e00552 100644
--- a/docs/communication.md
+++ b/docs/communication.md
@@ -1,7 +1,8 @@
 ---
 layout: default
 title: "Communication"
-nav_order: 5
+parent: Core Abstraction
+nav_order: 3
 ---
 
 # Communication
diff --git a/docs/flow.md b/docs/flow.md
index 1000ad4..7e2fcb1 100644
--- a/docs/flow.md
+++ b/docs/flow.md
@@ -1,7 +1,8 @@
 ---
 layout: default
 title: "Flow"
-nav_order: 4
+parent: Core Abstraction
+nav_order: 2
 ---
 
 # Flow
diff --git a/docs/index.md b/docs/index.md
index 6cfe5d7..a6b8913 100644
--- a/docs/index.md
+++ b/docs/index.md
@@ -23,6 +23,7 @@ We model the LLM workflow as a **Nested Flow**:
 ## Preparation
 
 - [LLM Integration](./llm.md)
+- [Tools](./tool.md)
 
 ## Core Abstraction
 
diff --git a/docs/llm.md b/docs/llm.md
index 7a0852f..c76c89e 100644
--- a/docs/llm.md
+++ b/docs/llm.md
@@ -1,14 +1,12 @@
 ---
 layout: default
 title: "LLM Integration"
-nav_order: 3
+nav_order: 2
 ---
 
 # LLM Wrappers  
 
-For your LLM app, implement a wrapper function to call LLMs yourself. 
-You can ask an assistant like ChatGPT or Claude to implement it. 
-For instance, asking ChatGPT to "implement a `call_llm` function that takes a prompt and returns the LLM response" gives:
+We **don't** provide built-in wrapper LLM wrappers. Instead, please implement your own, for example by asking an assistant like ChatGPT or Claude. If you ask ChatGPT to "implement a `call_llm` function that takes a prompt and returns the LLM response," you shall get something like:
 
 ```python
 def call_llm(prompt):
@@ -26,7 +24,7 @@ call_llm("How are you?")
 ```
 
 ## Improvements
-You can enhance the function as needed. Examples:
+Feel free to enhance your `call_llm` function as needed. Here are examples:
 
 - Handle chat history:
 
@@ -58,15 +56,15 @@ def call_llm(prompt):
 def call_llm(prompt):
     import logging
     logging.info(f"Prompt: {prompt}")
-    response = ...
+    response = ... # Your implementation here
     logging.info(f"Response: {response}")
     return response
 ```
 
 
-## Why not provide LLM Wrappers?
-I believe it is a bad practice to provide LLM-specific implementations in a general framework:
-- LLMs change frequently. Hardcoding them makes maintenance difficult.
-- You may need flexibility to switch vendors, use fine-tuned models, or deploy local LLMs.
-- You may need optimizations like prompt caching, request batching, or response streaming.
+## Why Not Provide a Built-in LLM Wrapper?
+I believe it is a **bad practice** to provide LLM-specific implementations in a general framework:
+- **LLM APIs change frequently**. Hardcoding them makes maintenance a nighmare.
+- You may need **flexibility** to switch vendors, use fine-tuned models, or deploy local LLMs.
+- You may need **optimizations** like prompt caching, request batching, or response streaming.
 
diff --git a/docs/node.md b/docs/node.md
index 22bc81b..022f235 100644
--- a/docs/node.md
+++ b/docs/node.md
@@ -1,7 +1,8 @@
 ---
 layout: default
 title: "Node"
-nav_order: 3
+parent: Core Abstraction
+nav_order: 1
 ---
 
 # Node