5 files changed, 56 insertions, 13 deletions
diff --git a/docs/docs/getting-started.md b/docs/docs/getting-started.md
index 0324f89e..bd77fe6e 100644
--- a/docs/docs/getting-started.md
+++ b/docs/docs/getting-started.md
@@ -2,13 +2,38 @@
 
 ## GitHub Codespaces Demo
 
-*TODO: Describe step-by-step how to try the GitHub Codespaces Demo*
+**TODO: Add `Open in GitHub Codespaces` badge here**
+
+1. Click the `Open in GitHub Codespaces` badge above
+
+:::info
+ We don't want to waste your time with install and env setup before you try Continue, so we set up a GitHub Codespaces dev container for you, which **won’t cost you anything**. If you are using GitHub Free for personal accounts, you can [use Codespaces for 120 hours per month for free](https://docs.github.com/en/billing/managing-billing-for-github-codespaces/about-billing-for-github-codespaces#monthly-included-storage-and-core-hours-for-personal-accounts)
+:::
+
+2. Select the `Create new codespace` button and wait a couple minutes for it to launch and then install the Continue extension. It should look like this one it is complete:
+
+**TODO: Insert an image of Continue when it has opened**
+
+3. Try playing with Continue as you build a Python script to do Pandas stuff
+
+**TODO: Design and set up Pandas stuff scenario in codespaces**
+
+4. There are a few recipes you should also try
+a. In the first directory, try out X recipe
+b. In the second directory, try out Y recipe
+c. In the third directory, try out Z recipe
+
+- database migrations
+- something super simple (libaries)
+- unit testing
+
+**TODO: Design and these recipes in codespaces**
 
 ## Next Steps
 
 If you would prefer to use Continue locally, we reccommend installing `Continue` packaged as a VS Code extension as described [here](./install.md).
 
-Otherwise, if you would like to continue to use GitHub Codespaces, then you should now go through the walkthroughs:
+Otherwise, if you would like to continue to use Continue on GitHub Codespaces, then you should now go through the walkthroughs:
 - How to [use the GUI](./walkthroughs/use-the-gui.md)
 - How to [use a recipe](./walkthroughs/use-a-recipe.md)
 - How to [create a recipe](./walkthroughs/create-a-recipe.md)
diff --git a/docs/docs/how-continue-works.md b/docs/docs/how-continue-works.md
index 7868c6be..751236fb 100644
--- a/docs/docs/how-continue-works.md
+++ b/docs/docs/how-continue-works.md
@@ -1,11 +1,9 @@
 # How `Continue` works
 
-_TODO: Describe in more detail how `Continue` works_
-
-> (Nate) Continue connects any code editor (primarily VS Code right now) to a server (the Continue server) that can take actions in the editor in accordance with defined recipes at the request of a user through the GUI. What this looks like (maybe I'll draw a diagram): The Continue VS Code extension runs the ContinueIdeProtocol, launches the Continue Python server in the background, and opens the Continue GUI in a side-panel. The Continue server is the brain, communication center, and source of truth, interacting with VS Code through the ContinueIdeProtocol and with the GUI through the NotebookProtocol. Communication between the extension and GUI happens through the Continue server. When you type a natural language command in the GUI, this is sent to the Continue server, where the `Autopilot` class takes action, potentially using the ContinueIdeProtocol to request actions be taken in the IDE, and then updates the GUI to display the new history.
-
 ![Continue Architecture Diagram](/img/continue-architecture.png)
 
+**TODO: Rename notebook protocol in this diagram**
+
 ## Overview
 
 The `Continue` library consists of a [SDK](./concepts/sdk.md), a [GUI](./concepts/gui.md), and a [Core](./concepts/core.md) that brings everything together.
@@ -18,6 +16,14 @@ The [Core](./concepts/core.md) connects the SDK and GUI with the IDE (i.e. in VS
 
 ## What to know about codebase
 
+**TODO: Refactor all of this and make it fit with language above**
+
+- Continue connects any code editor (primarily VS Code right now) to a server (the Continue server) that can take actions in the editor in accordance with defined recipes at the request of a user through the GUI
+- What this looks like:
+  - The Continue VS Code extension runs the ContinueIdeProtocol, launches the Continue Python server in the background, and opens the Continue GUI in a side-panel. 
+  - The Continue server is the brain, communication center, and source of truth, interacting with VS Code through the ContinueIdeProtocol and with the GUI through the NotebookProtocol. 
+  - Communication between the extension and GUI happens through the Continue server. 
+  - When you type a natural language command in the GUI, this is sent to the Continue server, where the `Autopilot` class takes action, potentially using the ContinueIdeProtocol to request actions be taken in the IDE, and then updates the GUI to display the new history.
 - `core` directory contains major concepts
   - This includes Autopilot, Policy, SDK (all in their own files so far)
   - It also includes `main.py`, which contains History, HistoryNode, Step, and others
@@ -30,4 +36,4 @@ The [Core](./concepts/core.md) connects the SDK and GUI with the IDE (i.e. in VS
 - `models` contains all the Pydantic models and `generate_json_schema.py`, a script that converts them to JSONSchema .json files in `schema/json`
 - `server` runs the servers that communicate with a) the React app (`notebook.py`) and b) the IDE (`ide.py`)
 - `ide_protocol.py` is just the abstract version of what is implemented in `ide.py`, and `main.py` runs both `notebook.py` and `ide.py` as a single FastAPI server. This is the entry point to the Continue server, and acts as a bridge between IDE and React app
-- We use OpenAPI/JSONSchema to define types so that it's really easy to bring them across language barriers. Use Pydantic types, then run `poetry run typegen` from the root of continuedev folder to generate JSONSchema json files in the `schema/json` folder. Then `npm run typegen` from the extension folder generates the types that are used within the extension.
+- We use OpenAPI/JSONSchema to define types so that it's really easy to bring them across language barriers. Use Pydantic types, then run `poetry run typegen` from the root of continuedev folder to generate JSONSchema json files in the `schema/json` folder. Then `npm run typegen` from the extension folder generates the types that are used within the extension.
+\ No newline at end of file
diff --git a/docs/docs/install.md b/docs/docs/install.md
index fdce84b0..f44fb01e 100644
--- a/docs/docs/install.md
+++ b/docs/docs/install.md
@@ -1,12 +1,22 @@
 # Installation
 
+:::note
 If you want to try `Continue` before installing, check out the [GitHub Codespaces Demo](./getting-started.md)
+:::
 
 ## Install `Continue` packaged as a VS Code extension
 
-Steps for installing `Continue` packaged as a VS Code extension...
+1. Click `Install` on the `Continue` extension in the Visual Studio Marketplace [here](https://marketplace.visualstudio.com/items?itemName=Continue.continue)
 
-*TODO: Describe step-by-step how to install `Continue` packaged as a VS Code extension*
+2. This will open the `Continue` extension page in VS Code, where you will need to click `Install` again
+
+3. Once you do this, you will see a message in the bottom right hand corner of VS Code that says `Setting up Continue extension...`. After a couple minutes, the `Continue` extension will then open up
+
+**TODO: Finish writing out this step-by-step**
+
+## How to install from source
+
+Please follow the [README instructions in the repo](https://github.com/continuedev/continue) to install `Continue` from source.
 
 ## Next steps
 
diff --git a/docs/docs/intro.md b/docs/docs/intro.md
index 2d92d777..e9f1dffe 100644
--- a/docs/docs/intro.md
+++ b/docs/docs/intro.md
@@ -2,15 +2,17 @@
 
 ![continue-cover-logo](/img/continue-cover-logo.png)
 
+**TODO: Nate's review of this page**
+
 ## What is `Continue`?
 
 **`Continue` is the open-source library for accelerating your use of LLMs while coding.**
 
-You define the scenarios where Large Language Models ([LLMs](./concepts/llm.md)) like GPT-4 and StarCoder should act as an autopilot that helps you complete software development tasks. You use [recipes](./concepts/recipe.md) created by others to automate more steps in your development workflows. If a [recipe](./concepts/recipe.md) does not exist or work exactly like you want, you can use the [Continue SDK](./concepts/sdk.md) to create custom [steps](./concepts/step.md) and compose them into personalized [recipes](./concepts/recipe.md). Whether you are using a [recipe](./concepts/recipe.md) created by yourself or someone else, you can also review, reverse, and rerun [steps](./concepts/step.md) with the [Continue GUI](./concepts/gui.md), which helps guide the work done by LLMs and learn when to use and trust them.
+You define the scenarios where Large Language Models ([LLMs](./concepts/llm.md)) like GPT-4 and StarCoder should act as an autopilot that helps you complete software development tasks. You use [recipes](./concepts/recipe.md) created by others to automate more steps in your development workflows. If a [recipe](./concepts/recipe.md) does not exist or work exactly like you want, you can use the [Continue SDK](./concepts/sdk.md) to create custom [steps](./concepts/step.md) and compose them into personalized [recipes](./concepts/recipe.md). Whether you are using a [recipe](./concepts/recipe.md) created by yourself or someone else, you can also review, reverse, and rerun [steps](./concepts/step.md) with the [Continue GUI](./concepts/gui.md), which helps you guide the work done by LLMs and learn when to use and trust them.
 
 ## Why do developers use `Continue`?
 
-Many developers have begun to use models like [GPT-4](https://openai.com/research/gpt-4) through [ChatGPT](https://openai.com/blog/chatgpt) while coding; however, this is quite a painful experience because of how much manual copy, paste, and editing is required to construct context for LLMs and then incorporate the generations from LLMs. Many other developers prefer to use open source models or work at organizations where they are unable to use ChatGPT, so they are using [StarCoder](https://huggingface.co/blog/starcoder) [Chat](https://huggingface.co/chat/) and running into the same issues.
+Many developers have begun to use models like [GPT-4](https://openai.com/research/gpt-4) through [ChatGPT](https://openai.com/blog/chatgpt) while coding; however, this is quite a painful experience because of how much manual copy, paste, and editing is required to construct context for LLMs and then incorporate the generations from LLMs. Many other developers prefer to use open source models or work at organizations where they are unable to use ChatGPT, so they are using [StarCoder](https://huggingface.co/blog/starcoder) [Chat](https://huggingface.co/chat/) and are running into the same issues.
 
 `Continue` eliminates the manual copy, paste, and editing required when using LLMs while coding. This accelerates how developers build, ship, and maintain software, while giving them the control to define when LLMs should take actions and the confidence to trust LLMs. In short, it enables developers to do what they have always done: work together to create better and better abstractions that make it easier and easier to automate the repetitive work that people want computers to do.
 
diff --git a/docs/docs/telemetry.md b/docs/docs/telemetry.md
index be0c7666..d1399706 100644
--- a/docs/docs/telemetry.md
+++ b/docs/docs/telemetry.md
@@ -8,10 +8,10 @@
 
 We track the following...
 
-*TODO: Detail exactly what we track*
+**TODO: Detail exactly what we track after we add initial telemetry**
 
 ## How to opt out
 
 Here are the instructions for turning off telemetry...
 
-*TODO: Describe step-by-step how to opt out of telemetry*
-\ No newline at end of file
+**TODO: Describe step-by-step how to opt out of telemetry**
+\ No newline at end of file