After a Ridgeline clinic uploads a new batch of post-op care PDFs to the Data Library, why must the content be indexed before an agent can ground answers in it?
Select an answer to reveal the explanation.
Short Explanation
Chunking cuts the document into pieces; indexing is what actually files those pieces so a retriever can find the right one fast instead of reading the whole library cover to cover every time someone asks a question.
Full Explanation
Indexing builds the searchable structure — including the embeddings vector search relies on — over the chunked content, so a retriever can quickly locate the passages most relevant to an incoming query instead of scanning raw text linearly. Without indexing, chunked content exists but isn't efficiently queryable, so grounding would be slow or unavailable. File-format conversion isn't what indexing does — the Data Library ingests common document formats without needing to transcode them into a different Salesforce-native format. Record ownership and sharing are governed by standard Salesforce security (profiles, permission sets, the Agent User's access), not by the indexing step, which is about search readiness rather than visibility control. Storage compression is also not the purpose here; indexing may have storage implications but its defining function is enabling retrieval, not shrinking files. For the exam: chunking segments content, indexing makes those segments searchable, and a retriever is the component that actually queries the index at run time.