Docker

Video Tutorial

1. EC2 Instance Setup and Configuration

1.1 EC2 Configuration

Follow these instructions to create a new EC2 instance with Ubuntu 24.04, t2.xlarge type, and 120 GiB gp3 storage.

Launch a new EC2 instance
In the “Application and OS Images” section, select Ubuntu
1. Look for 24.04 LTS in the list
In the “Instance Type” section, select t2.xlarge
In the “Configure Storage” section, set the size to 120 GiB, and make sure that “Volume Type” is set to gp3
Click “Launch Instance”

1.2 Set Inbound Security Rules

1.2.1. Accessing the Security Group Settings

Log into the AWS Management Console
Navigate to the EC2 Dashboard
Select your EC2 instance from the list
Click on the “Security” tab
Find the linked Security Group and click on it

1.2.2. Editing Inbound Rules

In the Security Group details, locate the “Inbound rules” section
Click the “Edit inbound rules” button

1.2.3. Adding New Rules

Add the following rules one by one using the “Add rule” button:

Port Range	Source
3001	0.0.0.0/0
80	0.0.0.0/0
3002	0.0.0.0/0
8000	0.0.0.0/0

Once complete, click “Save”.

1.3 Request Model Access on AWS Bedrock

1.3.1. Enable Amazon Bedrock

Sign in to the AWS Management Console.
Navigate to the Amazon Bedrock service.
If you haven’t used Bedrock before, you’ll see a “Get started” page.
Click on “Enable Amazon Bedrock”.
Review the terms and conditions, then click “Enable” to confirm.

1.3.2. Request Access to Sonnet 3.5 Model

In the Bedrock console, go to the “Model access” page.
Find “Sonnet 3.5” in the list of available models.
Click the checkbox next to “Sonnet 3.5”.
Click “Request model access” at the top of the page.
In the pop-up window, review the details and click “Submit request”.

1.3.3 Wait for Approval

Your request will be reviewed by AWS.
You’ll receive an email notification when your request is approved.

1.3.4 Verify Access

Once approved, return to the Bedrock console.
Go to the “Model access” page.
Confirm that “Sonnet 3.5” shows as “Access granted”.

2 Configuring Third-Party Applications

Communication Providers

Slack

Go to https://api.slack.com/apps
Click “Create New App” and choose “From an app manifest”
Select your workspace
Paste the App Manifest JSON from below into the provided field

{
"display_information": {
    "name": "duckie",
    "description": "A friendly developer support assistant :)",
    "background_color": "#7f807a"
},
"features": {
    "app_home": {
        "home_tab_enabled": false,
        "messages_tab_enabled": true,
        "messages_tab_read_only_enabled": false
    },
    "bot_user": {
        "display_name": "duckie",
        "always_online": true
    },
    "slash_commands": [
        {
            "command": "/duckie-response",
            "url": "http://[ec2_dns]:3002/slack/command/duckie-draft",
            "description": "Duckie Draft Response helps generate and improve responses to customers",
            "should_escape": false
        },
        {
            "command": "/duckie-investigate",
            "url": "http://[ec2_dns]:3002/slack/command/duckie-log-search",
            "description": "Search through and investigate your error logs to track down technical issues.",
            "should_escape": false
        },
        {
            "command": "/duckie-ask",
            "url": "http://[ec2_dns]:3002/slack/command/duckie-qa",
            "description": "Ask 🐥 a question!",
            "should_escape": false
        },
        {
            "command": "/duckie-resolution",
            "url": "http://[ec2_dns]:3002/slack/command/duckie-resolution",
            "description": "Work with 🐥 to resolve support issues!",
            "should_escape": false
        },
        {
            "command": "/duckie-track-ticket",
            "url": "http://[ec2_dns]:3002/slack/command/duckie-track-ticket",
            "description": "Track a ticket using Duckie",
            "should_escape": false
        },
        {
            "command": "/duckie-help",
            "url": "http://[ec2_dns]:3002/slack/command/duckie-help",
            "description": "Help",
            "should_escape": false
        }
    ]
},
"oauth_config": {
    "redirect_urls": [
        "http://[ec2_dns]:3001/client/slack/callback"
    ],
    "scopes": {
        "user": [
            "channels:read",
            "groups:read",
            "search:read"
        ],
        "bot": [
            "channels:history",
            "channels:manage",
            "channels:read",
            "chat:write",
            "files:read",
            "groups:history",
            "groups:read",
            "groups:write",
            "incoming-webhook",
            "mpim:read",
            "reactions:read",
            "reactions:write",
            "users:read",
            "im:read",
            "im:write",
            "im:history",
            "commands"
        ]
    }
},
"settings": {
    "event_subscriptions": {
        "request_url": "http://[ec2_dns]:3002/slack/events",
        "user_events": [
            "member_left_channel"
        ],
        "bot_events": [
            "app_uninstalled",
            "group_left",
            "member_joined_channel",
            "message.channels",
            "message.groups",
            "message.im",
            "reaction_added",
            "reaction_removed"
        ]
    },
    "interactivity": {
        "is_enabled": true,
        "request_url": "http://[ec2_dns]:3002/slack/interactivity"
    },
    "org_deploy_enabled": false,
    "socket_mode_enabled": false,
    "token_rotation_enabled": false
}
}

Update the values for URIs and URLs using your EC2 Public IPv4 DNS
Review the settings and click “Create”
Install the app to your workspace

Discord

Go to https://discord.com/developers/applications
Click “New Application” and name your app
In the “Bot” section, enable the bot and toggle on “Message Content Intent”
In “OAuth2 > General”:
- Add redirect: http://[ec2_dns]:3001/client/discord/callback
- Enable “Requires OAuth2 Code Grant”
In “OAuth2 > URL Generator”:
- Select “bot” scope
- Select these bot permissions:
  - Manage Webhooks
  - Read Messages/View Channels
  - Send Messages
  - Send Messages in Threads
  - Create Public Threads
  - Create Private Threads
  - Manage Messages
  - Manage Threads
  - Read Message History
Save changes

Ticketing Providers

Jira

Go to https://developer.atlassian.com/console/myapps/
Click “Create” and choose “OAuth 2.0 integration”
Name your app and click “Create”
In “Authorization”:
- Set callback URL: http://[ec2_dns]:3001/client/jira/callback
In “Permissions”:
- Add “Jira platform REST API” scopes:
  - read:jira-work
  - write:jira-work
  - manage:jira-webhook
- Add “OpenID Connect” scope:
  - offline_access
Save changes

Linear

Go to https://linear.app/duckie-demo/settings/api
Click “Create Key” and select “OAuth2 Application”
Fill in the details:
- Name your application
- Set Callback URL: http://[ec2_dns]:3001/client/linear/callback
Click “Create” and note your Client ID and Client Secret
In the “Webhooks” section:
- Enable webhooks
- Set Webhook URL: http://[ec2_dns]:3002/linear/webhook
- Select event types:
  - Issues
  - Comments
Click “Create Webhook”
Save all changes

Documentation Providers

Confluence

Go to https://developer.atlassian.com/console/myapps
Click “Create” and choose “OAuth 2.0 integration”
Name your app and click “Create”
In “Authorization”:
- Set callback URL: http://[ec2_dns]:3001/client/confluence/callback
In “Permissions”:
- Add “Confluence API” scopes:
  - read:content-details:confluence
  - read:page:confluence
  - write:page:confluence
  - read:content:confluence
  - read:attachment:confluence
  - read:comment:confluence
  - read:template:confluence
  - read:label:confluence
  - read:content.property:confluence
  - read:whiteboard:confluence
  - write:whiteboard:confluence
  - read:blogpost:confluence
  - write:blogpost:confluence
  - write:content:confluence
  - read:space-details:confluence
  - read:space:confluence
- Add “OpenID Connect” scope:
  - offline_access

Google Drive

Go to https://console.cloud.google.com/
Create a new project or select an existing one
Enable the Google Drive API for your project
Go to “Credentials” and click “Create Credentials” > “OAuth client ID”
Choose “Web application” as the application type
Set the authorized redirect URI: http://[ec2_dns]:3001/client/google/callback

3. Deploying Duckie on Your EC2 Instance

3.1 Setting Up Your Environment Variables

To configure your application correctly, you need to set up environment variables using a .env file. Follow these steps to create and populate your .env file on your EC2 instance.

Connect to your EC2 instance using SSH.
Navigate to the root directory of your application.
Create two new files named .backend.env and .frontend.env
Using the templates below, complete the .env files
You can find your 3rd party Client IDs where you created the applications

###############################################
###### 3rd Party Services Configurations ######
###############################################

## Slack Configuration
SLACK_CLIENT_ID=
SLACK_CLIENT_SECRET=
SLACK_BOT_OAUTH_TOKEN=

## Discord Configuration
DISCORD_CLIENT_ID=
DISCORD_CLIENT_SECRET=
DISCORD_BOT_OAUTH_TOKEN=

## Linear Configuration
LINEAR_CLIENT_ID=
LINEAR_CLIENT_SECRET=

## Confluence Configuration
CONFLUENCE_CLIENT_ID=
CONFLUENCE_CLIENT_SECRET=

## Jira Configuration
JIRA_CLIENT_ID=
JIRA_CLIENT_SECRET=

## Google Drive Configuration
GOOGLE_DRIVE_CLIENT_ID=
GOOGLE_DRIVE_CLIENT_SECRET=

## Sentry Configuration
SENTRY_CLIENT_ID=
SENTRY_CLIENT_SECRET=

# Zendesk Configuration
ZENDESK_CLIENT_ID=
ZENDESK_CLIENT_SECRET=

################################
###### LLM Configurations ######
################################
## You must configure one of the LLM services below

## AWS Bedrock Configuration
BEDROCK_AWS_REGION=
BEDROCK_AWS_KEY=
BEDROCK_AWS_SECRET_KEY=
BEDROCK_EMBEDDING_MODEL=amazon.titan-embed-text-v2:0 # DO NOT CHANGE

## OpenAI Configuration
OPENAI_API_KEY=

## Anthropic Configuration
ANTHROPIC_API_KEY=

##########################################
###### Data Services Configurations ######
##########################################

## Enable/Disable Local Data Services
WEAVIATE_LOCAL=True # DO NOT CHANGE
SUPABASE_AUTH_LOCAL=True # DO NOT CHANGE

## Weaviate Configuration
WEAVIATE_CLUSTER_URL=172.17.0.1 # DO NOT CHANGE
WEAVIATE_CLIENT_SECRET= # DO NOT CHANGE
WEAVIATE_LOCAL_PORT=8080 # DO NOT CHANGE

## Supabase Configuration - Used for storing app configurations and data
SUPABASE_URL=http://172.17.0.1:8000 # DO NOT CHANGE
SUPABASE_KEY=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyAgCiAgICAicm9sZSI6ICJzZXJ2aWNlX3JvbGUiLAogICAgImlzcyI6ICJzdXBhYmFzZS1kZW1vIiwKICAgICJpYXQiOiAxNjQxNzY5MjAwLAogICAgImV4cCI6IDE3OTk1MzU2MDAKfQ.DaYlNEoUrrEn2Ig7tqibS-PHK5vgusbcbo7X36XVt4Q # DO NOT CHANGE
SUPABASE_HOST=localhost # DO NOT CHANGE
SUPABASE_PORT=54322 # DO NOT CHANGE
SUPABASE_DB=postgres # DO NOT CHANGE
SUPABASE_USER=postgres # DO NOT CHANGE
SUPABASE_PASSWORD=postgres # DO NOT CHANGE

## Supabase Auth Configuration - Used for user authentication and authorization
SUPABASE_AUTH_URL=http://172.17.0.1:8000 # DO NOT CHANGE
SUPABASE_AUTH_KEY=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyAgCiAgICAicm9sZSI6ICJzZXJ2aWNlX3JvbGUiLAogICAgImlzcyI6ICJzdXBhYmFzZS1kZW1vIiwKICAgICJpYXQiOiAxNjQxNzY5MjAwLAogICAgImV4cCI6IDE3OTk1MzU2MDAKfQ.DaYlNEoUrrEn2Ig7tqibS-PHK5vgusbcbo7X36XVt4Q # DO NOT CHANGE

#######################################
###### Monitoring Configurations ######
#######################################
MONITORING=False
TRACING=False
EVALUATION=False

########################################
###### Environment Configurations ######
########################################
BACKEND_URL=http://[ec2_dns]:3002
FRONTEND_URL=http://[ec2_dns]:3001
HOST=0.0.0.0 # DO NOT CHANGE
PORT=3002 # DO NOT CHANGE
BINARY_BUILD=True # DO NOT CHANGE

##################################################
###### 3rd Party Integration Configurations ######
##################################################
REACT_APP_SLACK_CLIENT_ID=
REACT_APP_DISCORD_CLIENT_ID=
REACT_APP_JIRA_CLIENT_ID=
REACT_APP_LINEAR_CLIENT_ID=
REACT_APP_CONFLUENCE_CLIENT_ID=
REACT_APP_GOOGLE_DRIVE_CLIENT_ID=
REACT_APP_SENTRY_CLIENT_ID=
REACT_APP_ZENDESK_CLIENT_ID=

##########################################
###### Duckie Server Configurations ######
##########################################
REACT_APP_BACKEND_URL=http://[ec2_dns]:3002
REACT_APP_FRONTEND_URL=http://[ec2_dns]:3001

###################################
###### Client Configurations ######
###################################
HOST=0.0.0.0
PORT=3001 # DO NOT CHANGE

3.2 Setting Up Your GitHub Access Token for Cloning

3.2.1. Obtain Your Cloning Access Token

Reach out to the Duckie team for your cloning_access_token
You can contact us at founders@duckie.ai

3.2.2. Set Up the Token on Your EC2 Instance

Once you have received your cloning_access_token, follow these steps to set it up:

Connect to your EC2 instance using SSH.
Once connected, enter the following command, replacing your_cloning_access_token_here with the actual token you received:
```
export GITHUB_TOKEN=your_cloning_access_token_here
```
To verify that the token has been set correctly, you can echo the variable: This should display your token.
```
echo $GITHUB_TOKEN
```

4 Deploying Duckie on Your EC2 Instance

4.1 Prerequisites

Ensure you have sudo access on your EC2 instance.
Make sure you’ve set up your GitHub access token as described in the previous instructions.
Fully complete both .backend.env and .frontend.env

4.2 Running the Deployment Script

Connect to your EC2 instance via SSH.

Once connected, copy and paste the following command into your terminal:

curl -sSL https://duckie-public-dist.s3.amazonaws.com/duckie-selfhost-deploy.sh | sudo -E bash -s install

Press Enter to run the command
It will take ~2 minutes for Duckie to start

4.3 Upgrading Existing Installation

Connect to your EC2 instance via SSH.

Once connected, copy and paste the following command into your terminal:

curl -sSL https://duckie-public-dist.s3.amazonaws.com/duckie-selfhost-deploy.sh | sudo -E bash -s upgrade

Press Enter to run the command
It will take ~2 minutes for Duckie to redeploy

5. Start Using Duckie

Once all of the steps above have been complete, Duckie will be up and running in your AWS environment.

To use the Duckie web app, go to http://[ec2_dns]:3001.

To learn more about using Duckie, you can visit the Get Started page.

Getting Started

Using Duckie

Misc

Self Host Guide

Video Tutorial

1. EC2 Instance Setup and Configuration

1.1 EC2 Configuration

1.2 Set Inbound Security Rules

1.2.1. Accessing the Security Group Settings

1.2.2. Editing Inbound Rules

1.2.3. Adding New Rules

1.3 Request Model Access on AWS Bedrock

1.3.1. Enable Amazon Bedrock

1.3.2. Request Access to Sonnet 3.5 Model

1.3.3 Wait for Approval

1.3.4 Verify Access

2 Configuring Third-Party Applications

Communication Providers

Slack

Discord

Ticketing Providers

Jira

Linear

Documentation Providers

Confluence

Google Drive

3. Deploying Duckie on Your EC2 Instance

3.1 Setting Up Your Environment Variables

3.2 Setting Up Your GitHub Access Token for Cloning

3.2.1. Obtain Your Cloning Access Token

3.2.2. Set Up the Token on Your EC2 Instance

4 Deploying Duckie on Your EC2 Instance

4.1 Prerequisites

4.2 Running the Deployment Script

4.3 Upgrading Existing Installation

5. Start Using Duckie

Getting Started

Using Duckie

Misc

Self Host Guide

​Video Tutorial

​1. EC2 Instance Setup and Configuration

​1.1 EC2 Configuration

​1.2 Set Inbound Security Rules

​1.2.1. Accessing the Security Group Settings

​1.2.2. Editing Inbound Rules

​1.2.3. Adding New Rules

​1.3 Request Model Access on AWS Bedrock

​1.3.1. Enable Amazon Bedrock

​1.3.2. Request Access to Sonnet 3.5 Model

​1.3.3 Wait for Approval

​1.3.4 Verify Access

​2 Configuring Third-Party Applications

​Communication Providers

​Slack

​Discord

​Ticketing Providers

​Jira

​Linear

​Documentation Providers

​Confluence

​Google Drive

​3. Deploying Duckie on Your EC2 Instance

​3.1 Setting Up Your Environment Variables

​3.2 Setting Up Your GitHub Access Token for Cloning

​3.2.1. Obtain Your Cloning Access Token

​3.2.2. Set Up the Token on Your EC2 Instance

​4 Deploying Duckie on Your EC2 Instance

​4.1 Prerequisites

​4.2 Running the Deployment Script

​4.3 Upgrading Existing Installation

​5. Start Using Duckie

Video Tutorial

1. EC2 Instance Setup and Configuration

1.1 EC2 Configuration

1.2 Set Inbound Security Rules

1.2.1. Accessing the Security Group Settings

1.2.2. Editing Inbound Rules

1.2.3. Adding New Rules

1.3 Request Model Access on AWS Bedrock

1.3.1. Enable Amazon Bedrock

1.3.2. Request Access to Sonnet 3.5 Model

1.3.3 Wait for Approval

1.3.4 Verify Access

2 Configuring Third-Party Applications

Communication Providers

Slack

Discord

Ticketing Providers

Jira

Linear

Documentation Providers

Confluence

Google Drive

3. Deploying Duckie on Your EC2 Instance

3.1 Setting Up Your Environment Variables

3.2 Setting Up Your GitHub Access Token for Cloning

3.2.1. Obtain Your Cloning Access Token

3.2.2. Set Up the Token on Your EC2 Instance

4 Deploying Duckie on Your EC2 Instance

4.1 Prerequisites

4.2 Running the Deployment Script

4.3 Upgrading Existing Installation

5. Start Using Duckie