Sweep: Fix the documentation in docs/pages/deployment.mdx

[wwzeng1]

Details

Step 4 of the cloud development "update webhook URL" needs to be step 5 for the local instructions, updated with the ngrok URL

Branch

No response

Checklist

• Modify docs/pages/deployment.mdx ✓ b87f213 Edit

[sweep-nightly]

🚀 Here's the PR! #3725

See Sweep's progress at the progress dashboard!

💎 Sweep Pro: I'm using GPT-4. You have unlimited GPT-4 tickets. (tracking ID: 119948063f)

---

Tip

I can email you next time I complete a pull request if you set up your email here!

---

Actions (click)

• ↻ Restart Sweep

---

Step 1: 🔎 Searching

I found the following snippets in your repository. I will now analyze these snippets and come up with a plan.

Some code snippets I think are relevant in decreasing order of relevance (click to expand). If some file is missing from here, you can mention the path in the ticket description.

sweep/docs/pages/deployment.mdx

Lines 1 to 181 in d894442

	import { Callout, Tabs, Steps } from 'nextra/components'
	import { FaInfoCircle, FaExclamationCircle } from 'react-icons/fa'
	import YouTube from 'react-youtube';
	# Self-Hosting Sweep via Docker
	<Callout type="info">This is a guide for self-hosting Sweep. If you are interested in our hosted version please visit https://github.com/apps/sweep-ai.</Callout>
	Sweep GitHub App is an air-gapped, self-hostable version of Sweep which serves many repositories at once.<br/>
	Setting up the Sweep GitHub App involves:
	1. Creating the GitHub App and getting a trial license key.
	2. Hosting the Sweep backend on a cloud provider like DigitalOcean.
	**The trial license key is valid for 14 days.** To upgrade to an enterprise license key, contact us at [email protected]. Feel free to reach out to us for any help or questions.
	<Steps>
	### Create GitHub App
	Create a [GitHub App here](https://deploy.sweep.dev/) (1 minute). This tool will guide you through the process of creating a GitHub App, create a license key, and provide you with the necessary `.env` file.
	It will be named `.env.txt` and you will need to rename it to `.env` (browser security reasons). It should look something like this.
	```sh filename=".env"
	GITHUB_APP_ID=123456
	GITHUB_APP_PEM=-----BEGIN RSA PRIVATE KEY-----\n...\n-----END RSA PRIVATE KEY-----\n
	GITHUB_BOT_USER=sweepai-self-hosted
	LICENSE_KEY=ABCDE-FGHIJ-KLMNO-PQRST-UVWXY
	```
	---
	### Anthropic and OpenAI API keys
	Create an [OpenAI API key](https://platform.openai.com/account/api-keys) and an [Anthropic API key](https://console.anthropic.com/settings/keys) add it to your `.env`:
	```sh filename=".env"
	...
	ANTHROPIC_API_KEY=sk-ant-api03-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
	OPENAI_API_KEY=sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
	```
	---
	### Hosting Sweep
	To host Sweep you need to run the Sweep webhook on a publicly accessible server. We recommend using a cloud provider (DigitalOcean, AWS, or Azure) but you can also run it locally and use a reverse proxy like Ngrok.
	*If you encounter issues please reach out to us through [Email]([email protected]), [Discourse](https://community.sweep.dev/) or open a [GitHub issue](https://github.com/sweepai/sweep/issues). We are online, please ping us @kevin, @william, and @martin.*
	<Tabs items={["Cloud (recommended)", "Local"]}>
	<Tabs.Tab>
	We recommend deploying on the cloud for most users as the networking setup is easier. For this guide, we will use DigitalOcean.
	##### 1. Create Instance
	Recommended specs:
	- OS: Ubuntu
	- RAM: 32 GB
	- CPU: 4 vCPUs
	- Storage: 100 GB
	<details>
	<summary>If you do not have Docker/Docker Compose setup</summary>
	```sh filename="terminal"
	sudo apt install -y apt-transport-https ca-certificates curl software-properties-common -y
	curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo apt-key add -
	sudo add-apt-repository -y "deb [arch=amd64] https://download.docker.com/linux/ubuntu $(lsb_release -cs) stable"
	sudo apt update -y
	sudo apt install docker-ce -y
	sudo systemctl enable docker
	sudo systemctl start docker
	sudo usermod -aG docker ${USER}
	```
	</details>
	Start the docker client if you are on Mac/Windows.
	##### 2. Configure `.env` and pull Sweep
	```sh filename="terminal"
	git clone https://github.com/sweepai/sweep
	cd sweep
	vim .env
	```
	Copy the following from your local `.env` to your remote `.env`:
	```sh filename=".env"
	# Step 1
	GITHUB_APP_ID=123456
	GITHUB_APP_PEM=-----BEGIN RSA PRIVATE KEY-----\n...\n-----END RSA PRIVATE KEY-----\n
	GITHUB_BOT_USER=sweepai-self-hosted
	LICENSE_KEY=ABCDE-FGHIJ-KLMNO-PQRST-UVWXY
	# Step 2
	ANTHROPIC_API_KEY=sk-ant-api03-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
	OPENAI_API_KEY=sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
	```
	##### 3. Deploy the Sweep backend
	```sh filename="terminal"
	docker compose up --build hosted
	# To run in background, docker compose up --build -d hosted
	```
	##### 4. Update the GitHub Webhook URL
	Get the IP address of your instance and add the port Sweep is deployed to (in this case, 8080):
	```sh
	http://YOUR_IP:8080
	```
	Navigating to this link should look like this:
	{/* <img src="/deployment/digitalocean_step7.png" alt="Skip Resources Page" /> */}
	Go to your GitHub app and update `Webhook URL` to the link above.
	For user accounts:
	```txt /GITHUB_BOT_USERNAME/
	https://github.com/settings/apps/GITHUB_BOT_USERNAME
	```
	For organizations:
	```txt /GITHUB_BOT_USERNAME/, /YOUR_ORGANIZATION/
	https://github.com/YOUR_ORGANIZATION/sweepai/settings/apps/GITHUB_BOT_USERNAME
	```
	{/* <img src="/deployment/digitalocean_step9.png" alt="Skip Resources Page" /> */}
	</Tabs.Tab>
	<Tabs.Tab>
	If prefer to deploy on your local machine, you can do so by following the steps below.
	#### Prerequisites
	You need to install [Docker](https://docs.docker.com/engine/install/) to host Sweep.
	#### 1. Download the Docker Image
	Pull our image from Docker Hub:
	```sh filename="terminal"
	git clone https://github.com/sweepai/sweep
	cd sweep
	docker compose pull hosted
	```
	It may take 5-10 minutes to download the image and you can move on to the next step while it's downloading.
	#### 2. Set up the environment
	Create a `.env` file with the contents from steps 1 and 2.
	#### 3. Run the Docker Image
	Then in the same directory, run:
	```bash filename="terminal"
	docker compose up hosted # Add -d to run in background
	```
	#### 4. Set up the Reverse Proxy
	We're going to use [Ngrok](https://ngrok.com/) for a reverse proxy.
	Sign up for an Ngrok account and install the CLI from https://dashboard.ngrok.com/signup. Your terminal must be in the same directory as your ngrok installation. Start the reverse proxy with `ngrok http 8080`. The second last line should say something like
	```sh filename="terminal"
	Forwarding https://4d8d8bf053be.ngrok.app -> http://localhost:8080
	```
	Then proceed with step 4.
	</Tabs.Tab>
	</Tabs>
	</Steps>
	---
	### You have successfully deployed Sweep!
	*For enterprise support such as fine-tuning (20% performance gain), search index caching, usage tracking, and progress dashboards please contact [email protected].*
	**Creating a Pull Request with Sweep**
	Use Sweep by creating a new issue on the repository with a title prefixed with `Sweep:` like `Sweep: add type hints to BaseIndex.tsx`.
	For more details see [how to use Sweep](https://docs.sweep.dev/#-how-to-use-sweep).

---

Step 2: ⌨️ Coding

• Modify docs/pages/deployment.mdx ✓ b87f213 Edit

Modify docs/pages/deployment.mdx with contents:

Update the <original_code> and <new_code> blocks to match the actual code in the file and include the necessary changes.

<original_code>

4. Set up the Reverse Proxy

We're going to use Ngrok for a reverse proxy.

Sign up for an Ngrok account and install the CLI from https://dashboard.ngrok.com/signup. Your terminal must be in the same directory as your ngrok installation. Start the reverse proxy with ngrok http 8080. The second last line should say something like

Forwarding  https://4d8d8bf053be.ngrok.app -> http://localhost:8080

Then proceed with step 4.
</Tabs.Tab>

---

You have successfully deployed Sweep!

For enterprise support such as fine-tuning (20% performance gain), search index caching, usage tracking, and progress dashboards please contact [email protected].

Creating a Pull Request with Sweep

Use Sweep by creating a new issue on the repository with a title prefixed with Sweep: like Sweep: add type hints to BaseIndex.tsx.
For more details see how to use Sweep.
</original_code>

<new_code>

4. Set up the Reverse Proxy

We're going to use Ngrok for a reverse proxy.

Sign up for an Ngrok account and install the CLI from https://dashboard.ngrok.com/signup. Your terminal must be in the same directory as your ngrok installation. Start the reverse proxy with ngrok http 8080. The second last line should say something like

Forwarding  https://4d8d8bf053be.ngrok.app -> http://localhost:8080

5. Update the GitHub Webhook URL

Copy the forwarding URL from the ngrok output in step 4, for example:

https://4d8d8bf053be.ngrok.app

Go to your GitHub app and update the Webhook URL to the ngrok forwarding URL.

For user accounts:

https://github.com/settings/apps/GITHUB_BOT_USERNAME

For organizations:

https://github.com/YOUR_ORGANIZATION/sweepai/settings/apps/GITHUB_BOT_USERNAME

</Tabs.Tab>

---

You have successfully deployed Sweep!

For enterprise support such as fine-tuning (20% performance gain), search index caching, usage tracking, and progress dashboards please contact [email protected].

Creating a Pull Request with Sweep

Use Sweep by creating a new issue on the repository with a title prefixed with Sweep: like Sweep: add type hints to BaseIndex.tsx.
For more details see how to use Sweep.
</new_code>

---

Step 3: 🔁 Code Review

I have finished reviewing the code for completeness. I did not find errors for sweep/fix_the_documentation_in_docspagesdeploy.

---

🎉 Latest improvements to Sweep:

• New dashboard launched for real-time tracking of Sweep issues, covering all stages from search to coding.
• Integration of OpenAI's latest Assistant API for more efficient and reliable code planning and editing, improving speed by 3x.
• Use the GitHub issues extension for creating Sweep issues directly from your editor.

💡 To recreate the pull request edit the issue title or description.
^{Something wrong? Let us know.}

This is an automated message generated by Sweep AI.