Update docs.

website/blog/2024-11-15-CaptainAgent/index.mdx

@@ -4,6 +4,7 @@ authors:
   - jialeliu
   - LinxinS97
   - jieyuz2
+  - skzhang1
 tags: [LLM, GPT, AutoBuild]
 ---
 ![Illustration of how CaptainAgent build a team](img/overall.png)

website/docs/topics/captainagent/agent_library.mdx

@@ -1,8 +1,11 @@
 # Agent Library
+## Library Structure
 A simple agent in agent library requires three fields:
 - description: This describes the functionality of the agent.
 - system_message: This provides the system message of the agent for initialization.
 - name: The name of the agent.
+- model (optional): The backbone model of the agent.
+- tags (optional): The tag of the backbone model to use.
 
 An example of the agent library is as follows.
 ```
@@ -16,7 +19,7 @@ An example of the agent library is as follows.
 We provide a predefined agent library in `notebook/captainagent_expert_library.json`.
 
 ## Adding advanced agents
-We also support adding agents with advanced capability to the library, aside from agents with different system message. Just need to add a `model_path` field and any other arguments that needs to pass while initialization. For example, to add a WebSurferAgent:
+We also support adding agents with advanced capability to the library, aside from agents with customized system message. Just need to add a `agent_path` field and any other arguments that needs to pass while initialization. For example, to add a WebSurferAgent:
 
 ```
 [
@@ -37,3 +40,13 @@ We also support adding agents with advanced capability to the library, aside fro
     }
 ]
 ```
+
+Under the hood, the code will import the module according to the `agent_path`. The core code that implements this feature is:
+```
+module_path, model_class_name = agent_path.replace("/", ".").rsplit(".", 1)
+module = importlib.import_module(module_path)
+model_class = getattr(module, model_class_name)
+agent = model_class(**kwargs)
+```
+
+Make sure the correct path is provided to the config according to the code.