How It Works
1
Connect a Drive Account
Authenticate with Google OAuth using read-only Drive access. Each successful connection creates an installation representing that Google account.
2
Choose What to Index
Browse My Drive and shared drives, then pick specific files or folders. Folder selections recurse through all children, and Drive shortcuts resolve to their targets.
3
Search It Like Any Other Source
Once indexed, Drive content works with the same search, read, grep, and tree flows as other Nia sources.
4
Keep It Fresh with Sync
After the first full index, incremental sync updates only changed files instead of rebuilding everything.
Account Model
- Users connect Google Drive through Google OAuth with read-only Drive scopes.
- Each connection creates an installation tied to that Google account.
- Multiple Google accounts can be connected per user or organization.
Supported File Types
What Gets Indexed
Each selected file is downloaded, chunked, and embedded. Chunks preserve the original folder structure as file paths, and each chunk carries metadata such as:- file path
- modification time
- link back to the original Drive file
Search Experience
Semantic Search
Ask natural-language questions across all indexed Drive content or alongside your other Nia sources.
Read
Retrieve the content of specific Drive-backed files through the same source-reading flow as other indexed content.
Grep
Run regex pattern matching across indexed Drive files.
Tree
Browse the indexed file structure with the original folder hierarchy preserved.
Sync Behavior
After the initial full index, Nia tracks changes in Drive and re-indexes only affected files when content is modified, added, or deleted. Sync scopes are tracked separately for user change logs and shared drives so the system can keep Drive content current without rebuilding the entire source each time.API Surface
Google Drive content becomes searchable through the same unified search and source APIs used for repositories, docs, PDFs, datasets, Slack, and local sources.