Resources

Give the Agent the knowledge it needs to operate.

First, connect any external knowledge into Pylon through the Training Data ingestion process.

Training Data

By default, your Pylon knowledge base is always optionally available as a resource.

Pylon can:

• Crawl any public URL (tree of URLs) or ingest individual webpage URLs

• Ingest files uploads, like PDFs, CSVs, JSONs, etc.

Sources can be marked as public or internal depending on what workflows they should be used in. Individual AI Agents can be given access to different subsets of resources depending on use case.

Then, within your Agent configuration, grant access to the relevant resources.