Added explanations

Author: Dan Constantini

concepts/watcher.mdx

@@ -5,5 +5,5 @@ The -w (watch) command line option can be passed to the chainlit run command to
 When this option is specified, the file watcher will be started and any changes to files will cause the server to reload the app, allowing faster iterations.
 
 ```bash
-$ chainlit run app.py -w
+chainlit run app.py -w
 ```

examples/auto-gpt.mdx

@@ -2,9 +2,17 @@
 title: Auto GPT
 ---
 
-Example inspired from the [LangChain doc](https://python.langchain.com/en/latest/use_cases/autonomous_agents/autogpt.html)
+Auto-GPT is an AI tool that uses OpenAI's GPT-4 or GPT-3.5 APIs to perform autonomous tasks. It is an "AI agent" that can be given a goal in natural language and will attempt to achieve it by breaking it into sub-tasks and using the internet and other tools in an automatic loop. Unlike interactive systems such as ChatGPT, which require manual commands for every task, Auto-GPT assigns itself new objectives to work on with the aim of reaching a greater goal, without a mandatory need for human input.
 
-<Note>This example has extra dependencies, such as `openai`, `google-search-results` and `faiss-cpu`. You might have to install them manually.</Note>
+This example is inspired from the [LangChain doc](https://python.langchain.com/en/latest/use_cases/autonomous_agents/autogpt.html).
+
+<Note>This example has extra dependencies, such as `openai`, `google-search-results` and `faiss-cpu`.</Note>
+
+```bash
+pip install chainlit langchain openai google-search-results faiss-cpu
+```
+
+Then, you need to go to create an OpenAI key [here](https://platform.openai.com/account/api-keys) and get a SperpApi key [here](https://serpapi.com/).
 
 ## Code
 ```python autogpt.py
@@ -14,10 +22,20 @@ from langchain.tools.file_management.write import WriteFileTool
 from langchain.tools.file_management.read import ReadFileTool
 import os
 
+from langchain.vectorstores import FAISS
+from langchain.docstore import InMemoryDocstore
+from langchain.embeddings import OpenAIEmbeddings
+import faiss
+
+from langchain.experimental import AutoGPT
+from langchain.chat_models import ChatOpenAI
+import chainlit as cl
+
+
 os.environ["OPENAI_API_KEY"] = "OPENAI_API_KEY"
 os.environ["SERPAPI_API_KEY"] = "SERPAPI_API_KEY"
 
-
+# Tools the langchain agent will have access to
 search = SerpAPIWrapper()
 tools = [
     Tool(
@@ -29,23 +47,16 @@ tools = [
     ReadFileTool(),
 ]
 
-from langchain.vectorstores import FAISS
-from langchain.docstore import InMemoryDocstore
-from langchain.embeddings import OpenAIEmbeddings
 
 # Define your embedding model
 embeddings_model = OpenAIEmbeddings()
 # Initialize the vectorstore as empty
-import faiss
 
 embedding_size = 1536
 index = faiss.IndexFlatL2(embedding_size)
 
-from langchain.experimental import AutoGPT
-from langchain.chat_models import ChatOpenAI
-import chainlit as cl
-
 
+# Create the agent with the chainlit decorator
 @cl.langchain_factory
 def agent():
     vectorstore = FAISS(embeddings_model.embed_query, index, InMemoryDocstore({}), {})
@@ -66,10 +77,14 @@ def run(agent, input):
     return agent.run([input])
 ```
 
+Here you can see [langchain_run](/api-reference/langchain-run) and [langchain_factory](/api-reference/langchain-factory) were used. 
+Langchain agents usually take a string as input. The `AutoGPT` class takes in a list. These two helper functions allow to change the default behavior.  
+
+
 ## Try it out
 
 ```bash
-$ chainlit run autogpt.py
+chainlit run autogpt.py
 ```
 
 You can ask questions like `Create a weather report for san francisco`.

examples/mrkl.mdx

@@ -4,7 +4,11 @@ title: MRKL
 
 Example inspired from the [LangChain doc](https://python.langchain.com/en/latest/modules/agents/agents/examples/mrkl_chat.html)
 
-<Note>This example has extra dependencies, such as `google-search-results` and `openai`. You might have to install them manually.</Note>
+<Note>This example has extra dependencies, such as `google-search-results` and `openai`.</Note>
+
+```bash
+pip install chainlit langchain openai google-search-results
+```
 
 ## Code
 ```python mrkl.py
@@ -45,7 +49,7 @@ def load():
 ## Try it out
 
 ```bash
-$ chainlit run mrkl.py
+chainlit run mrkl.py
 ```
 
 You can ask questions like `What is the Paris weather forecast for tomorrow? How does it compare to today's?`.

examples/openai-sql.mdx

@@ -6,7 +6,11 @@ title: Text to SQL
 
 Let's build a simple agent that helps users to create SQL queries with natural language.
 
-<Note>This example has extra dependencies, such as `openai`. You might have to install them manually.</Note>
+<Note>This example has extra dependencies, such as `openai`.</Note>
+
+```bash
+pip install chainlit openai
+```
 
 ### Imports
 
@@ -67,7 +71,7 @@ def main(message: str):
 ### Try it out
 
 ```bash
-$ chainlit run app.py -w
+chainlit run app.py -w
 ```
 
 You can ask questions like `Compute the number of customers who watched more than 50 minutes of video this month`.

examples/qa.mdx

@@ -2,9 +2,19 @@
 title: Document QA
 ---
 
-Example inspired from the [LangChain doc](https://python.langchain.com/en/latest/use_cases/question_answering.html)
+In this example, we're going to build an chatbot QA app. In this example, we'll learn how to:
+- Upload a document
+- Create vector embeddings from a file
+- Create a chatbot app with the ability to display sources used to generate an answer
 
-<Note>This example has extra dependencies, such as `chromadb`, `tiktoken` and `openai`. You might have to install them manually.</Note>
+This example is inspired from the [LangChain doc](https://python.langchain.com/en/latest/use_cases/question_answering.html)
+
+<Note>This example has extra dependencies, such as `chromadb`, `tiktoken` and `openai`.</Note>
+```bash
+pip install chainlit langchain openai chromadb tiktoken
+```
+
+Then, you need to go to create an OpenAI key [here](https://platform.openai.com/account/api-keys) and get a SperpApi key [here](https://serpapi.com/).
 
 <Note>The state of the union file is available [here](https://github.com/hwchase17/langchain/blob/master/docs/modules/state_of_the_union.txt)</Note>
 
@@ -129,7 +139,7 @@ Then, we use [@langchain_postprocess](/api-reference/langchain-postprocess) (whi
 ## Try it out
 
 ```bash
-$ chainlit run qa.py
+chainlit run qa.py
 ```
 
 You can then upload any `.txt` file to the UI and ask questions about it.

installation.mdx

@@ -6,12 +6,12 @@ Chainlit requires `python>=3.8`.
 
 Open a terminal and run:
 ```bash
-$ pip install chainlit
+pip install chainlit
 ```
 
-To try if everything works correctly run:
+Make sure everything runs smoothly:
 ```bash
-$ chainlit hello
+chainlit hello
 ```
 
 This should spawn the chainlit UI and ask for your name like so:
@@ -24,5 +24,5 @@ This should spawn the chainlit UI and ask for your name like so:
 </Card>
 
 <Card title="With PurePython" icon="link" href="/pure-python">
-   Learn on how to use Chainlit with any python code.
+  Learn on how to use Chainlit with any python code.
 </Card>

langchain.mdx

@@ -26,7 +26,7 @@ def factory():
 That's it. Run the following command to start the chatbot UI.
 
 ```bash
-$ chainlit run app.py -w
+chainlit run app.py -w
 ```
 
 Congratulations, your first chainlit app is ready 🥳

mint.json

@@ -48,10 +48,10 @@
     {
       "group": "Examples",
       "pages": [
-        "examples/auto-gpt",
+        "examples/openai-sql",
         "examples/mrkl",
-        "examples/qa",
-        "examples/openai-sql"
+        "examples/auto-gpt",
+        "examples/qa"
       ]
     },
     {

overview.mdx

@@ -6,14 +6,16 @@ Chainlit is an open-source Python package that focuses on making it incredibly e
 
 
 ## Advantages of Chainlit
-1. **Enhanced explainability:** Chainlit's user interface enables you to visualize intermediary steps, enabling you to understand the reasoning behind an answer.
+1. **Chatbot UI:** Create cool LLM-based chatbot apps directly from your python code!
 
-2. **Debugging made easy:** Chainlit's prompt playground allows you to iterate on intermediate prompts, making it easier to identify and fix bugs.
+2. **Enhanced explainability:** Chainlit's user interface enables you to visualize intermediary steps, enabling you to understand the reasoning behind an answer.
 
-3. **Plug and play integration with LangChain:** Turn your LangChain agent in a chatbot UI with one line of code.
+3. **Debugging made easy:** Chainlit's prompt playground allows you to iterate on intermediate prompts, making it easier to identify and fix bugs.
 
-4. **Seamless file management:** The package includes a set of functions to ask and send files or documents.
+4. **Plug and play integration with LangChain:** Turn your LangChain agent in a chatbot UI with one line of code.
 
-5. **Free deployment:** Chainlit offers free deployment, allowing you to quickly and easily share your AI applications with your team or the world.
+5. **Seamless file management:** The package includes a set of functions to ask and send files or documents.
 
-6. **Improved team collaboration:** Chainlit's features, such as data persistence, human feedback, and prompt versioning, enable your team to work together more effectively and efficiently.
+6. **Deployment (coming soon):** Chainlit offers simple & free deployment, allowing you to quickly and easily share your AI applications with your team or the world.
+
+7. **Improved team collaboration (coming soon):** Chainlit's features, such as data persistence, human feedback, and prompt versioning, enable your team to work together more effectively and efficiently.

pure-python.mdx

@@ -21,7 +21,7 @@ def main(message: str):
 That's it. Run the following command to start the chatbot UI.
 
 ```bash
-$ chainlit run app.py -w
+chainlit run app.py -w
 ```
 
 Congratulations, your first chainlit app is ready 🥳