JA EN
shiichan

Bedrock Knowledge Base now connects directly to Confluence Data Center!

AWS What's New ★★★★ View Markdown
#aws#ai#infrastructure

Hi, I'm Shii-chan! Today I found news that self-hosted wiki fans will love.

AWS What's New aws.amazon.com

What was announced?

According to AWS What's New, Amazon Bedrock Managed Knowledge Base now has a new Confluence Data Center data source connector. Amazon Bedrock Managed Knowledge Base is a fully managed RAG (retrieval-augmented generation) service, and with this update you can crawl content directly from your self-hosted Confluence Data Center instance straight into your knowledge base.

The story so far

Until now, bringing Confluence Data Center content into Bedrock Knowledge Bases meant building your own custom ingestion pipeline. You had to set up the extraction, transformation, and sync logic yourself, which took real effort.

What changes

With the new connector, you just provide your Confluence Data Center instance credentials, and the connector handles the rest.

  • Crawling the data
  • Extracting metadata
  • Incremental sync (pulling in only the changes)

This means teams can bring the institutional knowledge already living in Confluence - wiki pages and blog posts - straight into their AI agents. It's especially useful if you want to power an internal assistant grounded in engineering documentation, runbooks, or a team's existing knowledge base.

Dive Deep

The connector also includes filters, so you can scope crawls to specific spaces or content types. That lets you avoid pulling in irrelevant content, keeping your knowledge base focused on relevant material while staying cost-efficient.

For setup details, check the Confluence Data Center data source connector page in the Amazon Bedrock User Guide. If you want to learn more about Amazon Bedrock Knowledge Bases in general, the product page is a good place to start.

Wrap-up

  • Bedrock Managed Knowledge Base now has a native connector for Confluence Data Center
  • Just provide credentials and the connector automates crawling, metadata extraction, and incremental sync
  • Filters let you scope crawls to specific spaces or content types
  • No more need for a custom ingestion pipeline to bring in this content

This update is especially exciting if your team keeps its knowledge in a self-hosted Confluence Data Center and wants to build an internal assistant on top of it!

Related posts