Skip to main content

Source and resources

Sources and resources are the two halves of how your data enters an AI system. A source is where data comes from, a connection to an external system such as a cloud bucket or a shared drive. A resource is what that source provides, an individual item such as a document, spreadsheet or image.

The relationship is simple: a source is a pipe, and resources are what flow through it. One source can supply many resources, and it keeps doing so over time as the underlying system changes.

Sources​

A source is a live connection to a system that holds your content, for example an Amazon S3 bucket or a Google Drive folder. Its job is to keep your data in sync: on a schedule, it checks the external system and brings new or changed items into the platform so your knowledge never goes stale. Sources are what make the difference between a one‑off import and continuously up‑to‑date knowledge.

Resources​

A resource is a single piece of content once it is inside the platform. Each resource carries more than its raw text:

  • Metadata such as a name, category and tags.
  • A link back to its original source and the time it was last synced.
  • The chunks and embeddings created from it during ingestion.
  • Relationships to other resources, which richer retrieval can follow.

Sources and resources in Rational AI​

You manage connections in the Sources section, where you add a source, set how often it synchronises and monitor each sync. The resources it brings in then live in the Knowledge module, where you can explore them, edit their metadata and see how they relate. Together they keep your agents working from current, well‑organised material.

Additional resources​