Skip to main content

AI Tooling Setup Guide

To get started using AI tooling at Open we primarily utilise AWS Bedrock which will require you to install the AWS CLI and create a profile that has access to Bedrock apis.

Prerequisites: AWS Setup

All tools require AWS CLI to be installed and configured. Complete these steps first:

1. Install AWS CLI

brew install awscli

2. Configure AWS CLI Profile

Create or update your AWS CLI configuration:

~/.aws/config
[sso-session OpenCloud]
sso_start_url = https://d-9767753c20.awsapps.com/start
sso_region = ap-southeast-2
sso_registration_scopes = sso:account:access

[profile op-rnd-admin]
sso_session = OpenCloud
sso_account_id = 592145485649
sso_role_name = AWSAdministratorAccess
region = ap-southeast-2

3. Authenticate

export AWS_PROFILE="op-rnd-admin"
aws sso login
# Test login
aws s3 ls
# 2024-10-15 09:23:19 bucket-123..
# 2024-10-15 09:23:19 open-bucket-abc..

Available Tools

ToolTypeModelBest For
ClaudeTUIAnthropic on BedrockComplex analysis & reasoning
OpenCodeCLIAny on BedrockCLI workflows & automation
KiroAppKiro/Amazon QQuick inline suggestions

Tool Installation & Setup

Select your preferred tool below to see installation and setup instructions:

Installation

brew install --cask claude-code

Setup

Configure Claude to use AWS Bedrock:

~/.claude/settings.json
{
"awsAuthRefresh": "aws sso login --profile op-rnd-admin",
"env": {
"AWS_PROFILE": "op-rnd-admin",
"AWS_REGION": "ap-southeast-2",
"CLAUDE_CODE_USE_BEDROCK": 1
}
}
GitLab View example ↗

Next Steps

  1. AWS Setup - You've completed the prerequisites
  2. 📦 Launch your tool - Start using it for your development work
  3. ⚙️ Add MCPs (optional) - Configure Model Context Protocol to give your tool access to your codebase

For detailed MCP setup and configuration for your specific tool, see the MCPs & Configuration guide.

Ready? Pick your tool from the tabs above and follow the installation and setup steps!