Instructions

1. Deploy IaC via CDK

cd ./cdk

# Create API key at https://app.pinecone.io/organizations/-/projects/-/keys
export PINECONE_API_KEY=

export CDK_APP_ID=
npx cdk deploy

2. Crawl data

Confluence

cd ./crawler
export OUTPUT_DIR=$(pwd)/crawled

# Create API token at https://id.atlassian.com/manage-profile/security/api-tokens
export CONFLUENCE_API_TOKEN=
export CONFLUENCE_EMAIL=
export CONFLUENCE_HOST=

npx tsx bin/confluence.ts id1 id2 id3

Upload to S3

Once all the data has been crawled, upload to S3 and start the ingestion job from UI (see below).

cd ./crawler
export OUTPUT_DIR=$(pwd)/crawled
aws s3 sync "${OUTPUT_DIR}" s3://${DocsBucketName} --delete

3. Build the UI

Create a .env.local file inside ./ui with these:

Group	Variables	Note
AWS access key	AWS_ACCESS_KEY_ID	AccessKeyId from CDK
	AWS_REGION	AwsRegion from CDK
	AWS_SECRET_ACCESS_KEY	AwsSecretAccessKey from CDK
Knowledge Base	KNOWLEDGE_BASE_ID	KnowledgeBaseId from CDK
	DATA_SOURCE_ID	DataSourceId from CDK
OpenAI	OPENAI_API_KEY	See OpenAI Platform

Then you can start the development server:

cd ./ui
npm run dev

Quick links:

• Claude 3.5 Sonnet via AWS Bedrock
• OpenAI
  • GPT-4o
  • GPT-4o Mini