diff --git a/connectors/bigquery/extractor.py b/connectors/bigquery/extractor.py
index 44ef386..5c8cc2b 100644
--- a/connectors/bigquery/extractor.py
+++ b/connectors/bigquery/extractor.py
@@ -428,11 +428,11 @@ def _init_extract_locked(
                     # Unverified entity type. Skip both the wrap view and
                     # the _meta row. The registry row remains; /api/v2/scan
                     # can still operate from it (builds BQ SQL from
-                    # bucket+source_table), and `da fetch` works.
+                    # bucket+source_table), and `agnes snapshot create` works.
                     logger.warning(
                         "Unverified BQ entity_type %r for %s.%s.%s — master view skipped. "
-                        "Use `da fetch` for this row, or file an issue with a repro to "
-                        "request native support.",
+                        "Use `agnes snapshot create` for this row, or file an issue with "
+                        "a repro to request native support.",
                         entity_type, project_id, dataset, source_table,
                     )
                     continue  # Do NOT insert _meta — no inner view to point at.
diff --git a/docs/DATA_SOURCES.md b/docs/DATA_SOURCES.md
index b7b1568..0f487db 100644
--- a/docs/DATA_SOURCES.md
+++ b/docs/DATA_SOURCES.md
@@ -74,7 +74,7 @@ bigquery:
 
 Registers BigQuery tables and views as remote DuckDB views (no data download). Queries
 issued through the master `analytics.duckdb` are forwarded to BigQuery via the DuckDB
-BigQuery extension. See also `da fetch` for the analytical workflow that materializes
+BigQuery extension. See also `agnes snapshot create` for the analytical workflow that materializes
 filtered subsets locally.
 
 ### Requirements
@@ -82,7 +82,7 @@ filtered subsets locally.
 - DuckDB BigQuery extension (auto-installed by the extractor on first run).
 - A GCP service account with `bigquery.metadata.get` on the dataset and
   `bigquery.data.viewer` (or finer) on the table; `bigquery.jobs.create` on the
-  billing project for views and `da fetch` queries.
+  billing project for views and `agnes snapshot create` queries.
 - Credentials resolution: GCE metadata server first, then Application Default
   Credentials (`gcloud auth application-default login` or
   `GOOGLE_APPLICATION_CREDENTIALS`). See `connectors/bigquery/auth.py`.
diff --git a/tests/test_cli_query_render.py b/tests/test_cli_query_render.py
index cc09eb8..cbe483a 100644
--- a/tests/test_cli_query_render.py
+++ b/tests/test_cli_query_render.py
@@ -4,7 +4,7 @@ Three CLI paths surface BQ errors today:
 - `agnes query --remote` (cli/commands/query.py:_query_remote → /api/query)
 - `agnes query --register-bq` (cli/commands/query.py:_query_hybrid via
   RemoteQueryError, which wraps server-side BqAccessError)
-- `da fetch` / `agnes schema` / etc. (cli/v2_client.V2ClientError → v2 endpoints)
+- `agnes snapshot create` / `agnes schema` / etc. (cli/v2_client.V2ClientError → v2 endpoints)
 
 After the refactor they all call cli.error_render.render_error so analyst
 output is consistent and structured. Closes part of #160 §4.7.3.
diff --git a/tests/test_cli_snapshot_create.py b/tests/test_cli_snapshot_create.py
index 9aa4eef..a03573c 100644
--- a/tests/test_cli_snapshot_create.py
+++ b/tests/test_cli_snapshot_create.py
@@ -1,4 +1,4 @@
-"""Tests for `agnes snapshot create` (folded from `da fetch`)."""
+"""Tests for `agnes snapshot create`."""
 
 from typer.testing import CliRunner