Skip to main content
What this builds. A KB that indexes one or more first-class Table SDK objects (already-stored Table resources on your account). You’ll end up with. A KB whose list_items surface includes the indexed tables as kind=TABLE items, queryable alongside text items. 
"""
Example 05: Index VectorShift Table objects into a KB.

`kb.add_tables(...)` accepts first-class VectorShift `Table` SDK
objects (already-stored Table resources, e.g. from a scrape-table flow).
CSV/XLSX/TSV file uploads go through `kb.add_files(...)` instead.
"""

from vectorshift import Table
from vectorshift.knowledge_base import KnowledgeBase

# In a real flow: fetch existing Table resources you want indexed.
# Replace the IDs with real Tables from your account.
table_ids = ["69f4b3e3b12ccdbaff50b0c9"]

tables = []
for tid in table_ids:
    if tid.startswith("YOUR_"):
        print("Replace YOUR_TABLE_ID_* with real Table ids to run this example.")
        raise SystemExit(0)
    tables.append(Table.fetch(id=tid))

kb = KnowledgeBase.new(name="SDK tables demo")
print(f"1. Created KB id={kb.id}")

task = kb.add_tables_and_wait(tables, timeout=180)
print(f"2. Indexed {len(tables)} tables: status={task.status}")

# Tables surface as KBItems with kind=TABLE
items = kb.list_items(limit=20)
print(f"3. Items: {[(i.name, i.kind) for i in items[:6]]}")

# kb.delete()
# print("4. Deleted KB.")

​
Expected output

1. Created KB id=...
2. Indexed 1 tables: status=...
3. Items: [('...', 'TABLE')]
For raw CSV/XLSX/TSV files, use add_files — add_tables is only for already-stored Table resources.

​
See also

Add files

Ingest CSV/XLSX/TSV files directly.

Advanced query

Query with filters, hybrid, rerank, and QA.

KB reference

Every public method.