SNOWFLAKE DOCS DIFF
総括は生成されていません（claude CLI が利用できなかった可能性があります）。
（この変更は要約対象外です。diff を参照してください）
判定根拠: 追加行に「has been removed」
--- ahttps://docs.snowflake.com/en/user-guide/cortex-code/changelog+++ bhttps://docs.snowflake.com/en/user-guide/cortex-code/changelog@@ -6,5 +6,24 @@ - [Cortex Code CLI](/user-guide/cortex-code/cortex-code-cli) -This page documents notable changes to Cortex Code CLI.+This page documents notable changes to CoCo CLI.++## 1.1.65 (2026-08-11)++### Added++- Data exploration subagent: CoCo CLI now includes a `data-discovery` subagent that can explore your Snowflake databases, schemas, and tables to answer questions about your data. It's enabled by default for all accounts. To turn it off, set `CORTEX_CODE_EXPERIMENTAL_FEATURES={"dataDiscovery":false}`.+- Managed settings enforcement: Administrators can enforce organization policies through a managed settings file. You can limit the Snowflake accounts and authentication methods that CoCo CLI is allowed to use and require a minimum CLI version. Use `cortex managed-settings` to inspect the active policy.+- Snowflake authentication for MCP servers: HTTP and SSE MCP servers can now authenticate with a named Snowflake connection. Add a `snowflake_connection` field to the server's entry in `mcp.json`, or use `cortex mcp add`, and CoCo CLI injects a fresh session token on every request. This gives you zero-configuration access to MCP servers hosted on Snowpark Container Services without a static token.+- Ignore ambient MCP servers: Added the `--only-explicit-mcp-servers` flag so a session loads only the MCP servers you pass in, for example with `--mcp-config`, and skips the servers configured in `mcp.json` and `~/.claude.json`.++### Changed++- `/clear` resets your conversation: `/clear` now starts a new conversation and clears the context instead of only clearing the screen. To clear just the screen, use the new `/cls` command or `Ctrl+L`. Your previous session is still available through `/resume`.+- Automatic VM sandbox management: When you use the VM sandbox, CoCo CLI now installs and maintains a version-matched sandbox for you. The `cortex update --with-vm-sandbox` flag has been removed because it's no longer needed.++### Fixed++- Symlink handling in file permission checks: File write and edit permission checks now resolve symbolic links before deciding whether a path is inside your workspace. Previously a symlink inside the workspace could point outside it and slip past the sandbox boundary.+- Typing lag: Fixed occasional input lag while typing in the CLI inside a Git repository, caused by a repeated `git` lookup on every keystroke. ## 1.1.60
（この変更は要約対象外です。diff を参照してください）
判定根拠: SQL 構文/コードブロックの増減 (+17/-0)
--- ahttps://docs.snowflake.com/en/developer-guide/streamlit/app-development/dependency-management+++ bhttps://docs.snowflake.com/en/developer-guide/streamlit/app-development/dependency-management@@ -7,5 +7,5 @@ dependencies in a `pyproject.toml` (recommended) or `requirements.txt` file. By default, your app doesn't have access to a package index like PyPI. Therefore, if you want to edit or specify the- versions of your app's dependencies, you must create an external access integration (EAI).+ versions of your app's dependencies, you must attach a Snowflake artifact repository. Additionally, you can install packages from wheel files included in your project directory. - Warehouse runtimes manage packages with [conda](https://docs.conda.io/en/latest/). You can@@ -32,6 +32,11 @@ </thead> <tbody>- <tr>- <td>PyPI or other external ["simple"](https://peps.python.org/pep-0503/) package indexes</td>+ <tr>+ <td>PyPI via Snowflake Artifact Repository (`snowflake.snowpark.pypi_shared_repository`)</td>+ <td>No</td>+ <td>Yes</td>+ </tr>+ <tr>+ <td>Other external ["simple"](https://peps.python.org/pep-0503/) package indexes</td> <td>No</td> <td>Yes (with EAI)</td>@@ -47,10 +52,6 @@ <td>Yes, but only via relative paths within the app's source files</td> </tr>- <tr>- <td>Snowflake Artifact Repository (`snowflake.snowpark.pypi_shared_repository`)</td>- <td>No</td>- <td>No</td>- </tr>- </tbody>++</tbody> </div>@@ -196,29 +197,131 @@ Not supported in government regions. -Container-runtime apps require an external access integration (EAI) to install packages-from an external package index like PyPI. Without an EAI, you can only use packages-shipped with the runtime or included in your app's source files.--Even if you only want to specify the version of Streamlit, you must include an EAI with-your app. Without an EAI, if you attempt to use version specifiers on pre-installed packages,-you might encounter an error when the runtime base image is updated. This is because your-version specifier might no longer be compatible with the pre-installed packages.--### External access integrations for container runtimes--For a general overview of external access integrations (EAIs), see [](/developer-guide/external-network-access/external-network-access-overview).--#### PyPI EAI- PyPI is the default package index used by uv to install Python packages in your container runtime.-Snowflake provides a managed network rule that simplifies creating an EAI for PyPI. Your account-administrator can use this rule to create a PyPI EAI and grant your role access to it. For setup-instructions, see [](#label-sis-pypi-eai-setup).--If you need to use a private or authenticated package repository such as JFrog Artifactory, your-administrator must create a custom EAI with the appropriate network rule and authentication secrets.-For an example, see [](#label-sis-jfrog-artifactory-example).--After your administrator has created a PyPI EAI and granted your role USAGE on it, you need to add+Container-runtime apps require a Snowflake artifact repository to install packages from an+external package index like PyPI. Without one, you can only use packages shipped with the runtime or included+in your app's source files.++Even if you only want to specify the version of Streamlit, you must attach an artifact+repository to your app. Without one, if you attempt to use version specifiers on pre-installed+packages, you might encounter an error when the runtime base image is updated. This is because+your version specifier might no longer be compatible with the pre-installed packages.++If an artifact repository cannot be used, you can configure external access integrations (EAIs)+(see [](/developer-guide/external-network-access/external-network-access-overview)) to access external+repositories directly. For more information see [](#label-sis-pypi-eai-setup).++<a id="label-streamlit-artifact-repository"></a>++### Snowflake artifact repository++Artifact repositories let your account administrator control which packages and versions+are available, without requiring outbound internet access. Snowflake provides a built-in shared PyPI mirror+(`snowflake.snowpark.pypi_shared_repository`). To learn more about the built-in mirror, see+[Artifact repository overview](/developer-guide/udf/python/udf-python-packages#artifact-repository-overview).++When one or more artifact repositories are attached to an app, packages are installed+exclusively from those repositories. Attaching or removing artifact repositories restarts+your app.++#### Prerequisites++Your role must have `USAGE` permission on the artifact repository:++- For Snowflake's built-in PyPI mirror:++ ```sql+ GRANT DATABASE ROLE SNOWFLAKE.PYPI_REPOSITORY_USER TO ROLE <your_role>;+ ```++- For a customer-hosted repository:++ ```sql+ GRANT USAGE ON ARTIFACT REPOSITORY <repo_name> TO ROLE <your_role>;+ ```++#### Attach an artifact repository++You can attach artifact repositories when creating an app, or add them to an existing app:++**Snowsight:**++Attach a repository when developing an app in [Workspaces](/developer-guide/streamlit/streamlit-in-workspaces/streamlit-in-workspaces-overview).++- For workspace development apps, open the **Settings** dialog and select the artifact+ repository from the dropdown.+- For deployed apps, select the artifact repository in the **Deploy** dialog.++After saving, the app restarts to fetch dependencies from the repository.++**SQL:**++When creating a new app:++```sql+CREATE STREAMLIT my_app+ RUNTIME_NAME = 'SYSTEM$ST_CONTAINER_RUNTIME_PY3_11'+ COMPUTE_POOL = my_compute_pool+ QUERY_WAREHOUSE = my_warehouse+ ARTIFACT_REPOSITORIES = (snowflake.snowpark.pypi_shared_repository);+```++To attach repositories to an existing app:++```sql+ALTER STREAMLIT my_app SET+ ARTIFACT_REPOSITORIES = (snowflake.snowpark.pypi_shared_repository, my_custom_repo);+```++To remove all artifact repositories from an app:++```sql+ALTER STREAMLIT my_app UNSET ARTIFACT_REPOSITORIES;+```++Attached repositories are visible in the output of `SHOW STREAMLITS`.++#### Customer-provided repositories++Your administrator can also create additional custom artifact repositories if they want more fine-grained control of package sources using managed+package indexes, like JFrog Artifactory. Using [customer-hosted repositories](/developer-guide/udf/python/customer-hosted-python-artifact-repositories) provides the following benefits:++- This helps prevent supply chain attacks and ensures packages come from trusted sources.+- It allows you to control which packages and versions are available to your apps.+- It provides audit trails for package installations.++<a id="label-sis-pypi-eai-setup"></a>++### External access integrations for repository access++The recommended way to access PyPI and other popular package sources is using [](#label-streamlit-artifact-repository).+If one or more artifact repositories are configured on an app, EAIs will not work as artifact repos+override the uv environment in the container runtime.++</Admonition >++In certain cases, you may need to access PyPI or another external repository, but cannot use an artifact repository+to do so. To facilitate that, Snowflake provides the ability to configure network access to those+external destinations via EAIs.++Snowflake provides a managed network rule, `SNOWFLAKE.EXTERNAL_ACCESS.PYPI_RULE`, that simplifies creating an+EAI for PyPI. Your account administrator can use this rule to create a PyPI EAI and grant your role access to it. The+following SQL commands create a PyPI EAI using the Snowflake-managed network rule and grant+USAGE to an app-development role:++```sql+USE ROLE ACCOUNTADMIN;++CREATE OR REPLACE EXTERNAL ACCESS INTEGRATION pypi_a... (truncated)
差分が長いため、途中まで表示しています。
（この変更は要約対象外です。diff を参照してください）
判定根拠: SQL 構文/コードブロックの増減 (+12/-6)
--- ahttps://docs.snowflake.com/en/user-guide/views-semantic/querying+++ bhttps://docs.snowflake.com/en/user-guide/views-semantic/querying@@ -477,6 +477,18 @@ ### Requirements for dimensions and metrics in a SELECT statement -In the SELECT statement, you can only refer to dimensions and metrics that have distinct names and that are not distinguished by-their logical table name. For example, suppose that a semantic view has two dimensions that have the unqualified name `name`:+When you specify the name of a semantic view in the `FROM` clause, you can reference dimensions, facts, and metrics+using their bare (unqualified) names or using dot-notation to qualify them with the entity (logical table) name.++**Bare names** work when a calculation name is unique across all entities in the semantic view, or when you reference+derived metrics and LOD metrics that are not bound to a specific entity:++```sql+SELECT customer_market_segment, AGG(order_average_value)+ FROM tpch_analysis+ GROUP BY customer_market_segment;+```++**Dot-notation** (`entity.calculation`) is required when two or more entities define a calculation with the same+name. For example, suppose that a semantic view has two dimensions that share the unqualified name `name`: ```sql@@ -487,6 +499,5 @@ ``` -In the SELECT statement, when you specify the qualified name of a dimension or metric, the qualifier is interpreted as the name-of the semantic view, not the name of a logical table:+Use dot-notation to specify which entity's calculation you want: ```sql@@ -496,8 +507,17 @@ ``` -```text-000904 (42000): SQL compilation error: error line 1 at position 7-invalid identifier 'NATION.NAME'-```+If you use a bare name that is ambiguous, the query fails with an error:++```sql+-- Fails: 'name' exists in both the nation and region entities+SELECT name FROM duplicate_names GROUP BY name;+```++```text+SQL compilation error: Ambiguous column name 'NAME'.+```++You can also use dot-notation for unambiguous calculations. In that case, the entity qualifier is optional but+can improve readability. <a id="label-semantic-views-query-standard-metrics"></a>@@ -648,10 +668,11 @@ ## Handling duplicate column names in the output -The output columns use the unqualified names of the metrics and dimensions. If you have multiple metrics and dimensions-with the same names, multiple columns will use the same name.--To work around this, use a table alias to assign different names to the columns.--For example, suppose that you define the following semantic view, which defines the dimensions `nation.name` and+When a semantic view contains multiple calculations with the same name across different entities, you can+use dot-notation (`entity.calculation`) to disambiguate them in both the `SEMANTIC_VIEW` clause and the+standard SQL `FROM` clause.++### Using dot-notation in standard SQL++For example, suppose that you define the following semantic view with the dimensions `nation.name` and `region.name`: @@ -674,11 +695,10 @@ ``` -If you query this view and select these two dimensions, the output includes two columns named `name` without any qualifiers:--```sql-SELECT * FROM SEMANTIC_VIEW(- duplicate_names- DIMENSIONS nation.name, region.name- );+Use dot-notation to select both dimensions:++```sql+SELECT nation.name, region.name+ FROM duplicate_names+ GROUP BY nation.name, region.name; ``` @@ -696,12 +716,10 @@ ``` -To disambiguate the columns, use a table alias to assign different column names (for example, `nation_name` and-`region_name`):--```sql-SELECT * FROM SEMANTIC_VIEW(- duplicate_names- DIMENSIONS nation.name, region.name- ) AS table_alias(nation_name, region_name);+To rename the output columns, use column aliases:++```sql+SELECT nation.name AS nation_name, region.name AS region_name+ FROM duplicate_names+ GROUP BY nation.name, region.name; ``` @@ -719,4 +737,18 @@ ``` +### SHOW COLUMNS behavior for ambiguous names++When a semantic view contains ambiguous calculation names, `SHOW COLUMNS` emits a separate row for+each calculation. For ambiguous names, the column name is reported using the `entity.calcName` convention+(for example, `orders.revenue`). This behavior ensures complete catalog visibility for both users and+downstream tools.++### BI tool compatibility++BI tools that rely on `SHOW COLUMNS` to discover available columns might automatically wrap dot-notation+column names in double quotes when generating queries (for example, `SELECT "orders.revenue" FROM sales_sv`).+Snowflake recognizes these quoted identifiers and resolves them to the correct entity and calculation,+so BI tool queries work without additional configuration.+ <a id="label-semantic-views-querying-window"></a>
（この変更は要約対象外です。diff を参照してください）
判定根拠: SQL 構文/コードブロックの増減 (+8/-0)
--- ahttps://docs.snowflake.com/en/user-guide/data-integration/openflow/connectors/oracle/setup-oracledb+++ bhttps://docs.snowflake.com/en/user-guide/data-integration/openflow/connectors/oracle/setup-oracledb@@ -40,5 +40,6 @@ 7. [](#label-set-xstream-outbound-server-connect-user) 8. [](#label-set-xstream-outbound-server-capture-user)-9. (Optional) [](#label-configure-ssl-connections)+9. (Optional) [](#label-oracle-standby-setup)+10. (Optional) [](#label-configure-ssl-connections) The steps in this topic are written for a multi-tenant architecture with a Container@@ -336,4 +337,10 @@ The XStream Outbound Server captures changes from redo logs for consumption by the Openflow Connector. Define which schemas or tables to replicate. For more information see [DBMS_XSTREAM_ADM.CREATE_OUTBOUND Documentation](https://docs.oracle.com/en/database/oracle/oracle-database/19/arpls/DBMS_XSTREAM_ADM.html#GUID-A602ED86-0F5A-4A27-92A0-55D5ADC0AF0D).++Create the outbound server on a writable database: the primary, a logical+standby, or a downstream capture database. You can't create it on an Active+Data Guard physical standby. For supported standby topologies, see+[](/user-guide/data-integration/openflow/connectors/oracle/about#label-oracle-standby-data-guard-support)+and [](#label-oracle-standby-setup). Important considerations for replication scope:@@ -472,4 +479,76 @@ in a single-tenant database. +<a id="label-oracle-standby-setup"></a>++## Data Guard or standby capture (optional)++By default, the procedures in this topic create the XStream outbound server on+the primary (source) database. If you want to keep replication load off the primary, review+[](/user-guide/data-integration/openflow/connectors/oracle/about#label-oracle-standby-data-guard-support)+and choose a supported topology before you create the outbound server.++<a id="label-oracle-logical-standby-requirements"></a>++### Logical standby requirements++On a logical standby, create the XStream outbound server and connect the+connector to that standby the same way you would for a primary. Before you+create the outbound server or start the connector, set Database Guard to+`STANDBY`.++Logical standbys default to Database Guard `ALL`, which blocks the XStream+client from reading the outbound server and raises+`ORA-16224: Database Guard is enabled`.++1. Check the current guard status:++ ```sql+ SELECT guard_status FROM v$database;+ ```++2. If the result is `ALL`, set Database Guard to `STANDBY`:++ ```sql+ ALTER DATABASE GUARD STANDBY;+ ```++Then complete the XStream setup procedures in this topic on the logical+standby, and point the connector's Oracle connection URL and XStream Out Server+URL at that standby.++### Active Data Guard requirements for snapshot load++You can run the connector's snapshot load against an Active Data Guard physical+standby to avoid reading application tables on the primary. Because the standby+is read-only:++- Set **Snapshot Fetching Strategy** to `SEQUENTIAL_BY_PRIMARY_KEY` in the+ connector parameters. `CONCURRENT_BY_ROWID` requires creating a parallel+ task, which isn't allowed on a read-only standby.+- Don't create the XStream outbound server on the Active Data Guard standby.+ For CDC, create it on the primary, a logical standby, or a downstream+ capture database instead.++Point the connector's Oracle connection URL at the Active Data Guard standby+for snapshot reads. Point the XStream Out Server URL at the writable database+that hosts the outbound server (for example, the primary or a downstream+capture database).++### Downstream capture for CDC++To run CDC off the primary, configure Oracle downstream capture so that a+separate database receives redo from the primary, then create the XStream+outbound server on that downstream database by using the procedures in this+topic.++Downstream capture provides redo only. It can't serve snapshot queries. If you+also need an initial snapshot without reading the primary, combine Active Data+Guard for the snapshot with downstream capture for CDC, or use a logical+standby for both stages.++For Oracle's downstream capture concepts and setup, see+[XStream Out Concepts](https://docs.oracle.com/en/database/oracle/oracle-database/19/xstrm/xstream-out-concepts.html)+in the Oracle documentation.+ <a id="label-configure-ssl-connections"></a>
（この変更は要約対象外です。diff を参照してください）
判定根拠: SQL 構文/コードブロックの増減 (+6/-0)
--- ahttps://docs.snowflake.com/en/sql-reference/sql/create-security-integration-oauth-snowflake+++ bhttps://docs.snowflake.com/en/sql-reference/sql/create-security-integration-oauth-snowflake@@ -51,4 +51,5 @@ [ OAUTH_ALTERNATE_REDIRECT_URIS = ( '<uri>' [ , '<uri>' , ... ] ) ] [ OAUTH_ENFORCE_PKCE = { TRUE | FALSE } ]+ [ OAUTH_ENABLE_ROLE_SELECTION = { TRUE | FALSE } ] [ OAUTH_SINGLE_USE_REFRESH_TOKENS_REQUIRED = { TRUE | FALSE } ] [ OAUTH_USE_SECONDARY_ROLES = { IMPLICIT | NONE } ]@@ -56,4 +57,5 @@ [ ALLOWED_ROLES_LIST = ( '<role_name>' [ , '<role_name>' , ... ] ) ] [ BLOCKED_ROLES_LIST = ( '<role_name>' [ , '<role_name>' , ... ] ) ]+ [ OAUTH_ANY_ROLE_MODE = { DISABLE | ENABLE | ENABLE_FOR_PRIVILEGE } ] [ OAUTH_ISSUE_REFRESH_TOKENS = { TRUE | FALSE } ] [ OAUTH_REFRESH_TOKEN_VALIDITY = <integer> ]@@ -313,4 +315,31 @@ </dd>+<dt>`OAUTH_ENABLE_ROLE_SELECTION = { TRUE | FALSE }`</dt>+<dd>++<div className="previewfeat sidebar">++<div className="sidebar-title">++%logo-snowflake-black% [Preview Feature](/release-notes/preview-features) — Open++</div>++Available to all accounts.++</div>++Controls whether the user can choose the role(s) for the session during the OAuth authorization flow, instead of the+integration using the user's default role. When `TRUE`, if the client does not request a specific role (that is, no+`session:role:role_name` scope), Snowflake presents a role-selection step during authorization where the user consents to one or more+roles or to all of their roles. The session is bound to the consented set: it can use only those roles, and role switching (with+<code className="samp">USE ROLE</code>) is limited to the consented roles. Selectable roles are bounded by the roles granted to the user+and the integration's `ALLOWED_ROLES_LIST` and `BLOCKED_ROLES_LIST`.++For more information, see [](/user-guide/oauth-custom#label-oauth-custom-role-selection).++Default: `FALSE`++</dd> <dt>`OAUTH_USE_SECONDARY_ROLES = { IMPLICIT | NONE }`</dt> <dd>@@ -356,4 +385,43 @@ Default: empty list (all roles other than those in `BLOCKED_ROLES_LIST` are allowed)++</dd>+<dt>`OAUTH_ANY_ROLE_MODE = { DISABLE | ENABLE | ENABLE_FOR_PRIVILEGE }`</dt>+<dd>++<div className="previewfeat sidebar">++<div className="sidebar-title">++%logo-snowflake-black% [Preview Feature](/release-notes/preview-features) — Open++</div>++Available to all accounts.++</div>++Controls whether a client or user can switch the session's primary role (using <code className="samp">USE ROLE <em>role</em>;</code>) to a role+other than the one bound to the OAuth access token. To switch roles, the client must also request the `session:role-any` scope in the+authorization request. Role switching is always restricted to roles granted to the user and permitted by `ALLOWED_ROLES_LIST` and+`BLOCKED_ROLES_LIST`.++- `DISABLE` does not allow the client or user to switch roles (i.e. <code className="samp">USE ROLE <em>role</em>;</code>). Default.+- `ENABLE` allows the client or user to switch roles.+- `ENABLE_FOR_PRIVILEGE` allows the client or user to switch roles only for a client or user with the `USE_ANY_ROLE`+ privilege on the integration. This privilege can be granted and revoked to one or more roles available to the user. For example:+ ```sql+ GRANT USE_ANY_ROLE ON INTEGRATION oauth_custom_1 TO role1;+ ```++ ```sql+ REVOKE USE_ANY_ROLE ON INTEGRATION oauth_custom_1 FROM role1;+ ```++This parameter applies only to the primary role of the session; secondary roles are controlled separately by+`OAUTH_USE_SECONDARY_ROLES`. You can only set this parameter when `OAUTH_CLIENT = CUSTOM`. The value can be optionally enclosed in+single quotes (e.g. either `DISABLE` or `'DISABLE'`).++Default: `DISABLE` </dd>
（この変更は要約対象外です。diff を参照してください）
判定根拠: SQL 構文/コードブロックの増減 (+0/-6)
--- ahttps://docs.snowflake.com/en/developer-guide/streamlit/object-management/security+++ bhttps://docs.snowflake.com/en/developer-guide/streamlit/object-management/security@@ -181,49 +181,4 @@ return the viewer's context. For more information, see [](/developer-guide/streamlit/features/restricted-callers-rights). -### Container runtimes only: Package repository access and security--This feature is not available in the People's Republic of China.--Not supported in government regions.--Configure one or more package indexes for container runtimes.--Container runtimes can install packages from external repositories like PyPI. You can-control package sources using managed package indexes, like JFrog Artifactory, or you can-use the default package index, PyPI. Regardless of which package index you use, you must-create an external access integration (EAI) to allow your apps to install dependencies.--Using a managed package index provides the following benefits:--- This helps prevent supply chain attacks and ensures packages come from trusted sources.-- It allows you to control which packages and versions are available to your apps.-- It provides audit trails for package installations.--For more information about how developers use EAIs to manage dependencies, see-[](#label-streamlit-managing-dependencies-container-runtimes). For more information-about setting up a managed package repository with authentication, see [](#label-sis-jfrog-artifactory-example).--<a id="label-sis-pypi-eai-setup"></a>--#### Set up a PyPI EAI for app developers--Container-runtime apps attempt to install dependencies from PyPI by default. Snowflake provides a managed network-rule, `SNOWFLAKE.EXTERNAL_ACCESS.PYPI_RULE`, that allows egress to PyPI. You can use this rule to-create a PyPI EAI without defining your own network rule. For more information about managed network rules,-see [](#label-snowflake-managed-egress-network-rules).--The following SQL commands create a PyPI EAI using the Snowflake-managed network rule and grant-USAGE to an app-development role:--```sql-USE ROLE ACCOUNTADMIN;--CREATE OR REPLACE EXTERNAL ACCESS INTEGRATION pypi_access_integration- ALLOWED_NETWORK_RULES = (snowflake.external_access.pypi_rule)- ENABLED = true;--GRANT USAGE ON INTEGRATION pypi_access_integration TO ROLE app_developer_role;-```- ### Warehouse runtimes only: External offerings terms
（この変更は要約対象外です。diff を参照してください）
判定根拠: 追加行に「deprecated」
--- ahttps://docs.snowflake.com/en/developer-guide/snowflake-app-runtime/getting-started+++ bhttps://docs.snowflake.com/en/developer-guide/snowflake-app-runtime/getting-started@@ -222,5 +222,6 @@ [account administrator setup](/developer-guide/snowflake-app-runtime/account-admin-setup). After setup, the file typically references the shared `SNOWFLAKE_APPS` database-and uses a **code stage** for uploaded source, for example:+and uses a **code workspace** for uploaded source when your role can create+workspaces. For example: ```yaml@@ -240,16 +241,24 @@ - node_modules - .env*+ - __pycache__+ - "*.pyc" - .next - .git+ - snowflake.log query_warehouse: SNOWFLAKE_APPS_QUERY_WH- code_stage: MY_APP_NAME_CODE-```+ code_workspace: SNOWFLAKE_APPS.PUBLIC.SNOWFLAKE_APPS+```++If your role can't create a workspace in the destination database, setup falls+back to a per-app **code stage** (for example `code_stage: MY_APP_NAME_CODE`)+instead. Setup resolves `database`, `schema`, and `query_warehouse` in this order: the-explicit `--database`, `--schema`, and `--warehouse` options, then Snowsight-account defaults from administrator setup, then your current connection-settings. After account setup, you normally don't pick these manually. The-`artifacts` section controls which files are uploaded, and the `ignore` list-excludes build output and local-only files.+explicit `--database`, `--schema`, and `--warehouse` options; then account+defaults from administrator setup; then the personal database default (when+account defaults aren't set); then your current connection settings. After+account setup, you normally don't pick these manually. The `artifacts` section+controls which files are uploaded, and the `ignore` list excludes build output+and local-only files. Run with `--dry-run` to preview the resolved configuration without writing@@ -321,5 +330,5 @@ If a deploy fails partway through, retry one phase with `--upload-only`,-`--build-only`, or `--deploy-only`. See+`--build-only`, or `--promote-only`. See [Retrying a failed deploy](#label-getting-started-retry-deploy). @@ -476,6 +485,9 @@ <td>`snow app events`</td> <td>- Fetch recent container logs (default: 500 lines). Use `--last N` for- more or fewer lines. Output is capped at 100 KB.+ Fetch observability streams for the Application Service. Default is+ recent container logs (`--type log`, 500 lines, capped at 100 KB). Use+ `--type metric` or `--type lifecycle` for event-table telemetry, and+ `--since` / `--until` for a time window. See+ [](/developer-guide/snowflake-app-runtime/observability). </td> </tr>@@ -502,6 +514,6 @@ - `snow app deploy --build-only`: re-trigger the build without re-uploading or redeploying.-- `snow app deploy --deploy-only`: create or upgrade the service without- re-uploading or rebuilding.+- `snow app deploy --promote-only`: create or upgrade the service without+ re-uploading or rebuilding. (`--deploy-only` is a deprecated alias.) Only one of these flags can be used at a time.
（この変更は要約対象外です。diff を参照してください）
判定根拠: SQL 構文/コードブロックの増減 (+2/-0)
--- ahttps://docs.snowflake.com/en/developer-guide/snowflake-app-runtime/observability+++ bhttps://docs.snowflake.com/en/developer-guide/snowflake-app-runtime/observability@@ -41,6 +41,8 @@ - `SUSPENDED`: the service is stopped and not billing compute. - `FAILED`: a container in the service encountered an unrecoverable error.+- `FAILING`: the service is transitioning to a failed state. - `DONE`: the service ran to completion (job-style runs only). - `CANCELLED`: a job-style run was cancelled.+- `CANCELLING`: a cancel is in progress. - `DELETING`: a drop is in progress. - `DELETED`: the service has been dropped.@@ -49,5 +51,5 @@ <a id="label-snowflake-app-runtime-observability-container-logs"></a> -## Read container logs+## Read container logs and telemetry The fastest way to read container logs is with the Snowflake CLI:@@ -59,7 +61,27 @@ By default, the command returns the 500 most recent log lines for the-Application Service defined in your `snowflake.yml`.+Application Service defined in your `snowflake.yml` (`--type log`). Output is+capped at 100 KB. -You can also read logs directly with SQL using+Use `--type` to select another observability stream:++```bash+# Historical logs from the event table (works after suspend; short ingestion lag)+snow app events --type log --since 6h++# CPU / memory / network metrics from the event table+snow app events --type metric --metric cpu --since 1h++# Service and container lifecycle events+snow app events --type lifecycle --since 2d+```++`--since` and `--until` accept relative shorthand such as `30m`, `6h`, or `2d`,+or an absolute UTC timestamp. Metric and lifecycle streams are historical and+default to the last hour when you don't pass a time window. With `--type+metric`, use `--metric cpu`, `memory`, or `network`, and `--raw` for+unconverted values.++You can also read live container logs directly with SQL using [SYSTEM$GET_APPLICATION_SERVICE_LOGS](/sql-reference/functions/system_get_application_service_logs). The function takes the fully qualified service name and, optionally, the number of tail lines to return (the
（この変更は要約対象外です。diff を参照してください）
判定根拠: SQL 構文/コードブロックの増減 (+1/-0)
--- ahttps://docs.snowflake.com/en/sql-reference/sql/alter-user+++ bhttps://docs.snowflake.com/en/sql-reference/sql/alter-user@@ -33,4 +33,6 @@ ALTER USER [ IF EXISTS ] [ <name> ] REMOVE DELEGATED { AUTHORIZATION OF ROLE <role_name> | AUTHORIZATIONS } FROM SECURITY INTEGRATION <integration_name>++ALTER USER [ IF EXISTS ] [ <name> ] REMOVE DELEGATED AUTHORIZATIONS OF ANY ROLE FROM SECURITY INTEGRATION <integration_name> ALTER USER [ IF EXISTS ] [ <name> ] mfaActions@@ -217,4 +219,27 @@ </dd>+<dt><code className="samp">REMOVE DELEGATED AUTHORIZATIONS OF ANY ROLE FROM SECURITY INTEGRATION <em>integration_name</em></code></dt>+<dd>++<div className="previewfeat sidebar">++<div className="sidebar-title">++%logo-snowflake-black% [Preview Feature](/release-notes/preview-features) — Open++</div>++Available to all accounts.++</div>++Revokes the user's consent to switch roles (the `session:role-any` scope) for the specified Snowflake OAuth security integration,+without affecting the user's consent for specific roles. This revokes any OAuth access token that relies on the any-role consent for+that integration.++Applies only to Snowflake OAuth custom-client integrations configured with `OAUTH_ANY_ROLE_MODE`. For more information, see+[](/user-guide/oauth-custom#label-oauth-custom-role-switching).++</dd> <dt><code className="samp">{ AUTHENTICATION | PASSWORD | SESSION } POLICY <em>policy_name</em> [ FORCE ]</code></dt> <dd>
（この変更は要約対象外です。diff を参照してください）
判定根拠: SQL 構文/コードブロックの増減 (+5/-0)
--- ahttps://docs.snowflake.com/en/user-guide/data-integration/openflow/connectors/oracle/troubleshoot+++ bhttps://docs.snowflake.com/en/user-guide/data-integration/openflow/connectors/oracle/troubleshoot@@ -388,4 +388,27 @@ - XStream has been created on this instance and has the same name. +## Error ORA-16224: Database Guard is enabled++Connecting to or reading from the XStream outbound server fails with:++```+oracle.streams.StreamsException: ORA-16224: Database Guard is enabled+```++This error occurs on a logical standby when Database Guard is set to `ALL`+(the default). With that setting, the XStream client can't read the outbound+server.++**Solution**++On the logical standby, set Database Guard to `STANDBY`:++```sql+ALTER DATABASE GUARD STANDBY;+```++For more information, see+[](/user-guide/data-integration/openflow/connectors/oracle/setup-oracledb#label-oracle-logical-standby-requirements).+ ## Error ORA-01722: invalid number when creating the outbound server
（この変更は要約対象外です。diff を参照してください）
判定根拠: 追加行に「deprecated」
--- ahttps://docs.snowflake.com/en/developer-guide/snowflake-app-runtime/deploy+++ bhttps://docs.snowflake.com/en/developer-guide/snowflake-app-runtime/deploy@@ -16,5 +16,6 @@ [Application Service](/sql-reference/sql/create-application-service) that serves your app at a stable live URL. Use `--upload-only`, `--build-only`, or-`--deploy-only` to run one phase at a time.+`--promote-only` to run one phase at a time. (`--deploy-only` is a deprecated+alias for `--promote-only`.) The build job and the Application Service are separate workloads with different@@ -67,5 +68,5 @@ <code>LATEST</code>). </td>- <td><code>--deploy-only</code></td>+ <td><code>--promote-only</code></td> </tr> </tbody>@@ -82,10 +83,14 @@ The upload phase copies project files named in your [`snowflake.yml`](/developer-guide/snowflake-cli/command-reference/snowflake-app-runtime-commands/snowflake-yml)-`artifacts` section to Snowflake storage. For a shared deploy destination,-uploaded source goes to an internal **stage** (`code_stage`). When the deploy-destination is your-[personal database](/user-guide/personal-databases), Snowflake uses a-**workspace** (`code_workspace`) instead, because personal databases don't-support stages.+`artifacts` section to Snowflake storage. When your `snowflake.yml` sets+`code_workspace`, uploaded source goes to that **workspace**. When it sets+`code_stage` instead, uploaded source goes to an internal **stage**. Personal+databases don't support stages, so PDB deploys use a workspace.++`snow app setup` usually writes `code_workspace` as the default. For a standard+database, it falls back to `code_stage` only when your role can't create a+workspace in the destination schema. Specifying an existing `code_workspace` in+`snowflake.yml` doesn't require that create privilege by itself; you need access+to write to that workspace at deploy time. The `ignore` list in `snowflake.yml` excludes local-only paths such as
（この変更は要約対象外です。diff を参照してください）
判定根拠: SQL 構文/コードブロックの増減 (+3/-0)
--- ahttps://docs.snowflake.com/en/collaboration/provider-listings-auto-fulfillment-troubleshoot-setup+++ bhttps://docs.snowflake.com/en/collaboration/provider-listings-auto-fulfillment-troubleshoot-setup@@ -157,5 +157,11 @@ Explore the cost ramifications for auto-fulfilling a database larger than 10TB to one or more regions. See [](/collaboration/provider-understand-cost-auto-fulfillment). -If you accept the potential added cost, you can contact [Snowflake Support](https://docs.snowflake.com/user-guide/contacting-support) to have the limit adjusted for your entire account.+If you accept the potential added cost, an account administrator (a user with the ACCOUNTADMIN role) can raise the limit for your account by setting the [](/sql-reference/parameters#label-listing-auto-fulfillment-initial-refresh-size-limit-in-tb) parameter. For example:++```sql+ALTER ACCOUNT SET LISTING_AUTO_FULFILLMENT_INITIAL_REFRESH_SIZE_LIMIT_IN_TB = 20.0;+```++To remove the size limit, set the value to `0.0`. </dd>
（この変更は要約対象外です。diff を参照してください）
変更内容: 本文を更新（105行追加・103行削除）
--- ahttps://docs.snowflake.com/en/user-guide/cortex-code/cli-reference+++ bhttps://docs.snowflake.com/en/user-guide/cortex-code/cli-reference@@ -21,21 +21,24 @@ ## CLI options -| Option | Description |-| ------------------------------------ | --------------------------------------------------------- |-| `-c, --connection <name>` | Use specific Snowflake connection |-| `-w, --workdir <path>` | Set working directory for file operations |-| `-m, --model <model_name>` | Specify AI model to use |-| `--plan` | Plan mode: require approval before all actions |-| `--bypass` | Automatically approve all planned actions |-| `--dangerously-allow-all-tool-calls` | Disable tool call permission prompts (caution) |-| `--continue` | Resume most recent conversation |-| `-r, --resume <session_id>` | Resume specific session by ID, or `last` for last session |-| `-p, --print "<prompt>"` | Pass specified prompt, print response, and exit |-| `-f, --file <file>` | Read prompt from file, execute, and exit |-| `--output-format stream-json` | JSON output (for scripting) |-| `-V, --version` | Show installed version |-| `--help` | Show CLI help |--Connections must be defined in `~/.snowflake/connections.toml`. See [](/user-guide/cortex-code/cortex-code-cli) for connection setup. Session IDs are shown at startup, at exit, and stored in `~/.snowflake/cortex/conversations/`.+| Option | Description |+| ------------------------------------------------ | -------------------------------------------------------------------- |+| `-c, --connection <name>` | Use specific Snowflake connection |+| `-w, --workdir <path>` | Set working directory for file operations |+| `-m, --model <model_name>` | Specify AI model to use |+| `--plan` | Plan mode: require approval before all actions |+| `--bypass`, `--dangerously-allow-all-tool-calls` | Enable bypass mode, automatically approving all tool calls |+| `--continue` | Resume most recent conversation |+| `-r, --resume [<session_id>]` | Resume specific session by ID, or `last` for the most recent session |+| `-p, --print "<prompt>"` | Pass specified prompt, print response, and exit |+| `--output-format stream-json` | JSON output (for scripting) |+| `-V, --version` | Show installed version |+| `--help` | Show CLI help |++Connections must be defined in `~/.snowflake/connections.toml`. See [](/user-guide/cortex-code/cortex-code-cli) for connection setup. Session IDs are shown at startup and at exit, and they are stored in `~/.snowflake/cortex/conversations/`.++`--bypass` and `--dangerously-allow-all-tool-calls` are aliases of the same option. Omit the value of `-r, --resume` to open the+resume picker instead of naming a session.++Bypass mode approves every tool call without prompting you first. Use it only when you trust every action the agent might take. ### Examples@@ -43,5 +46,5 @@ Start with working directory: -```bash+```shell cortex -w /path/to/project ```@@ -49,5 +52,5 @@ Resume last session with specific connection: -```bash+```shell cortex --continue -c production ```@@ -55,5 +58,5 @@ One-off prompt (JSON output): -```bash+```shell cortex -p "List all Python files" --output-format stream-json ```@@ -68,11 +71,36 @@ | `cortex --version` | Verify after update | +### `exec`++Use `exec` to run CoCo CLI non-interactively, which is useful in continuous integration and continuous delivery (CI/CD)+pipelines. Plan mode is always disabled and interactive prompts are automatically rejected.++| Command | Description |+| --------------------------- | ----------------------------------- |+| `cortex exec "<prompt>"` | Run a prompt and exit |+| `cortex exec --file <file>` | Read the prompt from a file |+| `cortex exec --file -` | Read the prompt from standard input |++<a id="label-coco-cli-mcp-subcommands"></a>+ ### `mcp` -| Command | Description |-| --------------------------------- | ---------------------------- |-| `cortex mcp list` | List configured servers |-| `cortex mcp add` | Add new server (interactive) |-| `cortex mcp remove <server_name>` | Remove server |+| Command | Description |+| -------------------------------------------------- | ---------------------------------------------------------------- |+| `cortex mcp list` | List configured servers |+| `cortex mcp add <name> <command_or_url> [args...]` | Add new server |+| `cortex mcp get <name>` | Show details for a configured server |+| `cortex mcp remove <name>` | Remove server |+| `cortex mcp reconnect [name]` | Reconnect one server, or all servers if you don't provide a name |+| `cortex mcp start` | Start configured servers |++The `add` subcommand accepts the following options:++| Option | Description |+| -------------------------------- | ---------------------------------------------------------------------- |+| `-t, --transport <type>` | Transport to use: `stdio`, `sse`, or `http`. Default: `stdio`. |+| `-e, --env <KEY=value>` | Set an environment variable. Repeat the option for multiple variables. |+| `-H, --header "<name>: <value>"` | Set an HTTP header. Repeat the option for multiple headers. |+| `--timeout <milliseconds>` | Connection timeout. | See [MCP (Model Context Protocol)](extensibility#extensibility-mcp) for details.@@ -126,14 +154,10 @@ | ---------------------------- | ------------------------------------------------------- | | `/help` | Show interactive help |-| `/plan` | Enable planning mode |-| `/plan_off` | Disable planning mode | | `/clear`, `/cls` | Clear the screen | | `/new` | Start a new session | | `/rename <title>` | Rename current session ... (truncated)
差分が長いため、途中まで表示しています。
（この変更は要約対象外です。diff を参照してください）
変更内容: 本文を更新（103行追加・7行削除）
--- ahttps://docs.snowflake.com/en/user-guide/oauth-custom+++ bhttps://docs.snowflake.com/en/user-guide/oauth-custom@@ -16,11 +16,6 @@ The optional "scope" parameters in the initial authorization request limit the role permitted by the access token and can additionally- be used to configure the refresh token behavior.--In-session role switching to secondary roles is not supported with Snowflake OAuth.--If this behavior is necessary with your OAuth workflow, use External OAuth instead.--For more information, see [](#label-ext-oauth-secondary-roles).+ be used to configure the refresh token behavior. Clients can request the `session:role-any` scope when they need [in-session role switching](#label-oauth-custom-role-switching). Activation of secondary roles ([USE SECONDARY ROLES](/sql-reference/sql/use-secondary-roles)) is controlled separately by the+ OAUTH_USE_SECONDARY_ROLES parameter. ## Create a Snowflake OAuth integration@@ -57,4 +52,86 @@ sessions, and you can't combine them. `IMPLICIT` auto-activates the user's default secondary roles, which would bypass the allowlist, so you can only set ALLOWED_ROLES_LIST when `OAUTH_USE_SECONDARY_ROLES = NONE` (the default).++<a id="label-oauth-custom-role-switching"></a>++### Allowing clients to switch roles++<div className="previewfeat sidebar">++<div className="sidebar-title">++%logo-snowflake-black% [Preview Feature](/release-notes/preview-features) — Open++</div>++Available to all accounts.++</div>++Snowflake OAuth sessions are single-role by default: [USE ROLE](/sql-reference/sql/use-role) is blocked unless you enable role switching on the integration. Set OAUTH_ANY_ROLE_MODE in [CREATE SECURITY INTEGRATION](/sql-reference/sql/create-security-integration-oauth-snowflake) or [ALTER SECURITY INTEGRATION](/sql-reference/sql/alter-security-integration-oauth-snowflake).++- `DISABLE` (default): the session can't switch roles.+- `ENABLE`: the session can switch roles.+- `ENABLE_FOR_PRIVILEGE`: the session can switch roles only if the user holds the `USE_ANY_ROLE` privilege on the integration (through+ any role granted to the user). This privilege determines whether the session can switch roles at all; it doesn't restrict which+ roles the session can switch to.++In addition to enabling the mode, the client must request the `session:role-any` scope in the authorization request (see+[Scope](#label-oauth-custom-scope)). A session can only switch to roles that are granted to the user and permitted by+ALLOWED_ROLES_LIST and BLOCKED_ROLES_LIST.++If the integration also uses role selection (`OAUTH_ENABLE_ROLE_SELECTION`), the roles the user consents to further bound switching:+the session can switch only among the consented roles. When the user consents to all roles — or the integration doesn't use role+selection — switching is bounded only by the user's grants and the allow/block lists.++Snowflake re-checks the integration's role-switching configuration each time the session switches roles, so disabling role switching+(setting OAUTH_ANY_ROLE_MODE to DISABLE) or revoking a required privilege prevents further switches in existing sessions.++The OAUTH_ANY_ROLE_MODE controls the session's primary role. Activation of secondary roles+([USE SECONDARY ROLES](/sql-reference/sql/use-secondary-roles)) is controlled separately by the OAUTH_USE_SECONDARY_ROLES parameter+and the user's consent.++To revoke a user's consent to switch roles for an integration, use+[ALTER USER … REMOVE DELEGATED AUTHORIZATIONS OF ANY ROLE](/sql-reference/sql/alter-user).+<a id="label-oauth-custom-role-selection"></a>++### Letting users select roles during authorization++<div className="previewfeat sidebar">++<div className="sidebar-title">++%logo-snowflake-black% [Preview Feature](/release-notes/preview-features) — Open++</div>++Available to all accounts.++</div>++When a client omits a `session:role:role_name` scope in the authorization requests, the session uses the user's+default role. To let the user pick the session role(s) during the OAuth authorization flow, set+OAUTH_ENABLE_ROLE_SELECTION parameter to TRUE when you create the integration (using+[CREATE SECURITY INTEGRATION](/sql-reference/sql/create-security-integration-oauth-snowflake)) or later (using+[ALTER SECURITY INTEGRATION](/sql-reference/sql/alter-security-integration-oauth-snowflake)).++When OAUTH_ENABLE_ROLE_SELECTION is TRUE, Snowflake presents a role-selection step during authorization. The user can select and consents to one or+more roles — or to all of their roles — and the session is bound to the consented set: the session can use only those roles, and role+switching with [USE ROLE](/sql-reference/sql/use-role) is limited to the roles the user consented to. If the user consents to all+roles, the session can use any role granted to the user.++The roles offered for selection are bounded by the roles granted to the user and by the integration's ALLOWED_ROLES_LIST and+BLOCKED_ROLES_LIST. When a user consents to more than one role, the consented roles form the session's role universe: they are available both as+the session's primary role (selectable with USE ROLE) and as its secondary roles (activated with USE SECONDARY ROLES). If a user+consents to a single role, the session is fixed to that role: it can't switch roles with USE ROLE and can't activate secondary roles+with USE SECONDARY ROLES.++When role selection is in effect, the consented roles govern the session's secondary roles, and the OAUTH_USE_SECONDARY_ROLES+parameter does not apply. Changing OAUTH_USE_SECONDARY_ROLES does not affect a session that was already created from a role-selection+consent.++Consent is additive. Re-authorizing expands a user's consented set of roles; it never removes roles, and consenting to all roles+persists until revoked. To reduce or reset a user's consented roles, the user or an administrator must revoke the delegated+authorization with [ALTER USER … REMOVE DELEGATED AUTHORIZATIONS](/sql-reference/sql/alter-user). ### Supporting multiple redirect URIs@@ -304,4 +381,6 @@ </blockquote> +<a id="label-oauth-custom-scope"></a>+ ##### Scope @@ -357,4 +436,21 @@ `AUTH SNOWFLAKE` (with a space), then the value of the `scope` query parameter must be `session:role-encoded:AUTH%20SNOWFLAKE`.++</td>+ </tr>+ <tr>+ <td>`session:role-any`</td>+ <td>No</td>+ <td>++Requests a session that can switch its primary role at runtime (using [USE ROLE](/sql-reference/sql/use-role)) instead of being fixed+to a single role. The user consents to this capability, and the session can then switch to any role granted to the user that is+permitted by the integration's `ALLOWED_ROLES_LIST` and `BLOCKED_ROLES_LIST`.++This scope is honored only when the security integration sets `OAUTH_ANY_ROLE_MODE` to `ENABLE` or `ENABLE_FOR_PRIVILEGE`; otherwise+the authorization request is rejected. You can combine this scope with a `session:role:role_name` scope to select the initial primary+role. If you omit the role scope, the user's default role is used.++For more information, see [](#label-oauth-custom-role-switching). </td>
（この変更は要約対象外です。diff を参照してください）
変更内容: 本文を更新（56行追加・46行削除）
--- ahttps://docs.snowflake.com/en/user-guide/cortex-code/keyboard-shortcuts+++ bhttps://docs.snowflake.com/en/user-guide/cortex-code/keyboard-shortcuts@@ -10,51 +10,54 @@ ## Input shortcuts -| Shortcut | Action |-| ---------- | ----------------------------------------------- |-| Enter | Submit message |-| Ctrl-J | Insert newline (multiline input) |-| Ctrl/Cmd-V | Paste from clipboard |-| Ctrl-C | Cancel/interrupt (double-tap to exit) |-| Esc | Dismiss suggestions (double-tap to clear input) |-| Ctrl-K | Kill line (clear input) |-| Ctrl-Y | Yank (paste killed text) |-| Ctrl-A | Move to line start |-| Ctrl-E | Move to line end |+| Shortcut | Action |+| ---------- | --------------------------------------------------------------------------- |+| Enter | Submit message |+| Ctrl-J | Insert newline (multiline input) |+| Ctrl/Cmd-V | Paste from clipboard |+| Ctrl-C | Cancel/interrupt (double-tap to exit) |+| Esc | Cancel the current action or dismiss the active overlay (context-dependent) |+| Ctrl-K | Kill line (delete from cursor to end of line) |+| Ctrl-Y | Yank (paste killed text) |+| Ctrl-A | Move to line start |+| Ctrl-E | Move to line end |+| Ctrl-B | Move cursor left |+| Ctrl-F | Move cursor right |+| Ctrl-D | Delete character under cursor | ## View shortcuts -| Shortcut | Action |-| ------------ | --------------------------------- |-| Ctrl-T | Open table viewer (cycle forward) |-| Ctrl-Shift-T | Cycle table viewer backward |-| Ctrl-P | Toggle compact/expanded mode |-| Ctrl-O | Open full transcript viewer |-| Ctrl-B | View background bash processes |-| Ctrl-D | Open todo/task viewer |-| Ctrl-G | Open web search results |-| ? | Toggle help overlay |+| Shortcut | Action |+| -------- | -------------------------------------------------------------------------------------- |+| Ctrl-T | Open table viewer (cycle forward through cached SQL query results) |+| Ctrl-O | Cycle display mode: compact → expanded → transcript |+| Alt-T | Open or close the fullscreen todo viewer |+| Alt-G | Open fullscreen web search results view |+| Ctrl-S | Open or close the background agent (subagent) picker |+| Ctrl-B | If a bash tool call is running, move it to the background; otherwise go back one level |+| ? | Toggle help overlay (only when the input line is empty) | -## Mode Shortcuts+## Mode shortcuts -| Shortcut | Action |-| --------- | ----------------------- |-| Shift-Tab | Cycle operational modes |+| Shortcut | Action |+| --------- | ------------------------------------------------------------ |+| Ctrl-P | Toggle plan mode |+| Ctrl-G | Toggle team mode |+| Shift-Tab | Cycle permission level (confirm actions ↔ bypass safeguards) | -Modes cycle in this order:--- Confirm actions-- Plan mode-- Bypass safeguards+Plan mode (Ctrl-P) and team mode (Ctrl-G) are independent toggles. They are not steps in the Shift-Tab permission-level cycle. ## History Navigation -| Shortcut | Action |-| ----------- | ---------------------------------------------- |-| Ctrl-R | Reverse search history (Emacs-style) |-| Ctrl-S | Forward search history |-| Ctrl-G | Cancel history search |-| Up / Down | Navigate history (when at top/bottom of input) |-| Option-Up | Previous history entry |-| Option-Down | Next history entry |+| Shortcut | Action |+| ----------- | -------------------------------------------------------------- |+| Ctrl-R | Open history search; press again for the next older match |+| Ctrl-S | While history search is open, go to the previous (newer) match |+| Ctrl-G | Cancel history search |+| Esc | Cancel history search |+| Up / Down | Navigate history (when at top/bottom of input) |+| Option-Up | Previous history entry |+| Option-Down | Next history entry |++Ctrl-S and Ctrl-G are context-dependent: outside of history search, Ctrl-S opens the background agent picker and Ctrl-G toggles team mode. ### Table viewer shortcuts@@ -83,5 +86,5 @@ ## Display modes -Press Ctrl-P to toggle between Compact and Expanded display modes for tool execution details.+Press Ctrl-O to cycle through Compact, Expanded, and Transcript display modes for tool execution details. <dl>@@ -100,4 +103,10 @@ </dd>+<dt>Transcript mode</dt>+<dd>++- Full scrollable conversation transcript++</dd> </dl> @@ -110,8 +119,8 @@ │ INPUT │ VIEW │ │ Enter Submit │ Ctrl-T Table viewer │-│ Ctrl-J Newline │ Ctrl-P Toggle compact │-│ Ctrl-C Cancel │ Ctrl-O Transcript │-│ Ctrl-R History │ Ctrl-B Background bash │-│ Esc Esc Clear │ Ctrl-D Todo viewer │+│ Ctrl-J Newline │ Ctrl-O Cycle display mode │+│ Ctrl-C Cancel │ Alt-T Todo viewer │+│ Ctrl-R History │ Alt-G Web search results │+│ Esc Cancel │ Ctrl-S Subagent picker │ │ │ ? Help overlay │ ├─────────────────────────┼────────────────────────────────┤@@ -123,5 +132,6 @@ │ q/Esc Exit │ Tab Accept │ ├─────────────────────────┴────────────────────────────────┤-│ Shift-Tab Cycle modes (Confirm → Plan → Bypass) │+│ Shift-Tab Cycle permission level (confirm ↔ bypass) │+│ Ctrl-P Toggle plan mode Ctrl-G Toggle team mode │ └──────────────────────────────────────────────────────────┘ ```@@ -129,6 +139,6 @@ ## Tips -- **Use Ctrl-P often**: Switch between compact and expanded mode based on what you need to see.+- **Use Ctrl-O often**: Switch between compact, expanded, and transcript display modes based on what you need to see. - **Master Ctrl-R**: History search is powerful for repeating complex prompts. - **Vim-style navigation**: The h/j/k/l keys work everywhere for cursor movement.-- **Double-tap patterns**: For example, Esc Esc clears input, Ctrl-C Ctrl-C exits.+- **... (truncated)
差分が長いため、途中まで表示しています。
（この変更は要約対象外です。diff を参照してください）
変更内容: 本文を更新（57行追加・0行削除）
--- ahttps://docs.snowflake.com/en/sql-reference/parameters+++ bhttps://docs.snowflake.com/en/sql-reference/parameters@@ -172,4 +172,8 @@ <td>[INITIAL_REPLICATION_SIZE_LIMIT_IN_TB](#initial-replication-size-limit-in-tb)</td> <td></td>+ </tr>+ <tr>+ <td>[LISTING_AUTO_FULFILLMENT_INITIAL_REFRESH_SIZE_LIMIT_IN_TB](#listing-auto-fulfillment-initial-refresh-size-limit-in-tb)</td>+ <td>Used to set the maximum size limit for the initial refresh of a listing data product for Cross-Cloud Auto-Fulfillment.</td> </tr> <tr>@@ -5162,4 +5166,57 @@ For examples of how to use this parameter, see [](#label-fetching-integer-data-types-as-bigint). +<a id="label-listing-auto-fulfillment-initial-refresh-size-limit-in-tb"></a>++## LISTING_AUTO_FULFILLMENT_INITIAL_REFRESH_SIZE_LIMIT_IN_TB++<div className="newfeat sidebar">++<div className="sidebar-title">++Supported regions for feature++</div>++This feature is not available in the People's Republic of China.++</div>++<dl className="field-list">+<dt>Type<span className="colon">:</span></dt>+<dd>++Account — Can be set only for Account++</dd>+<dt>Data Type<span className="colon">:</span></dt>+<dd>++Number.++</dd>+<dt>Description<span className="colon">:</span></dt>+<dd>++Sets the maximum estimated size limit for the initial refresh of a listing data product for Cross-Cloud Auto-Fulfillment (in TB). Set this parameter on the provider account. This size limit helps prevent accounts from accidentally incurring large auto-fulfillment charges.++To remove the size limit, set the value to `0.0`.++Note that there is currently no default size limit applied to subsequent refreshes of a data product.++</dd>+<dt>Values<span className="colon">:</span></dt>+<dd>++`0.0` and above with a scale of at least 1 (e.g. `20.5`, `32.25`, `33.333`, etc.).++</dd>+<dt>Default<span className="colon">:</span></dt>+<dd>++`10.0`++</dd>+</dl>+ <a id="label-listing-auto-fulfillment-replication-refresh-schedule"></a>
（この変更は要約対象外です。diff を参照してください）
変更内容: 本文を更新（56行追加・0行削除）
--- ahttps://docs.snowflake.com/en/sql-reference/sql/alter-security-integration-oauth-snowflake+++ bhttps://docs.snowflake.com/en/sql-reference/sql/alter-security-integration-oauth-snowflake@@ -62,7 +62,9 @@ [ OAUTH_ALTERNATE_REDIRECT_URIS = ( '<uri>' [ , '<uri>' , ... ] ) ] [ OAUTH_ENFORCE_PKCE = { TRUE | FALSE } ]+ [ OAUTH_ENABLE_ROLE_SELECTION = { TRUE | FALSE } ] [ PRE_AUTHORIZED_ROLES_LIST = ( '<role_name>' [ , '<role_name>' , ... ] ) ] [ ALLOWED_ROLES_LIST = ( '<role_name>' [ , '<role_name>' , ... ] ) ] [ BLOCKED_ROLES_LIST = ( '<role_name>' [ , '<role_name>' , ... ] ) ]+ [ OAUTH_ANY_ROLE_MODE = { DISABLE | ENABLE | ENABLE_FOR_PRIVILEGE } ] [ OAUTH_ISSUE_REFRESH_TOKENS = { TRUE | FALSE } ] [ OAUTH_REFRESH_TOKEN_VALIDITY = <integer> ]@@ -328,4 +330,28 @@ </dd>+<dt>`OAUTH_ENABLE_ROLE_SELECTION = { TRUE | FALSE }`</dt>+<dd>++<div className="previewfeat sidebar">++<div className="sidebar-title">++%logo-snowflake-black% [Preview Feature](/release-notes/preview-features) — Open++</div>++Available to all accounts.++</div>++Specifies whether the user can choose the role or roles for the session during the OAuth authorization flow, instead of the+integration using the user's default role. When enabled, the session is bound to the consented set of roles. Selectable roles are+bounded by the roles granted to the user and the integration's `ALLOWED_ROLES_LIST` and `BLOCKED_ROLES_LIST`. For more information, see+[CREATE SECURITY INTEGRATION (Snowflake OAuth)](/sql-reference/sql/create-security-integration-oauth-snowflake) and+[](/user-guide/oauth-custom#label-oauth-custom-role-selection).++Default: `FALSE`++</dd> <dt>`OAUTH_USE_SECONDARY_ROLES = { IMPLICIT | NONE }`</dt> <dd>@@ -370,4 +396,34 @@ Default: empty list (all roles other than those in `BLOCKED_ROLES_LIST` are allowed)++</dd>+<dt>`OAUTH_ANY_ROLE_MODE = { DISABLE | ENABLE | ENABLE_FOR_PRIVILEGE }`</dt>+<dd>++<div className="previewfeat sidebar">++<div className="sidebar-title">++%logo-snowflake-black% [Preview Feature](/release-notes/preview-features) — Open++</div>++Available to all accounts.++</div>++Specifies whether a client or user can switch the session's primary role (using <code className="samp">USE ROLE <em>role</em>;</code>) to a role+other than the one bound to the OAuth access token. To switch roles, the client must also request the `session:role-any` scope in the+authorization request. Role switching is always restricted to roles granted to the user and permitted by `ALLOWED_ROLES_LIST` and+`BLOCKED_ROLES_LIST`.++- `DISABLE` does not allow the client or user to switch roles. Default.+- `ENABLE` allows the client or user to switch roles.+- `ENABLE_FOR_PRIVILEGE` allows the client or user to switch roles only for a client or user with the `USE_ANY_ROLE` privilege on+ the integration.++This parameter applies only to the primary role of the session; secondary roles are controlled separately by+`OAUTH_USE_SECONDARY_ROLES`. You can only set this parameter when `OAUTH_CLIENT = CUSTOM`. For more information, see+[CREATE SECURITY INTEGRATION (Snowflake OAuth)](/sql-reference/sql/create-security-integration-oauth-snowflake). </dd>
（この変更は要約対象外です。diff を参照してください）
変更内容: 本文を更新（53行追加・0行削除）
--- ahttps://docs.snowflake.com/en/user-guide/data-integration/openflow/version-history+++ bhttps://docs.snowflake.com/en/user-guide/data-integration/openflow/version-history@@ -350,4 +350,57 @@ - Salesforce: Added parameter to enable detailed per-object record operation metrics (created, updated, deleted) for the merge query. +## August 11, 2026++<a id="label-control-plane-core-0-128-0"></a>++### Control Plane Core 0.128.0++- Security patches and dependency upgrades.++<a id="label-data-plane-service-0-128-0"></a>++### Data Plane Service 0.128.0++- Security patches and dependency upgrades.++<a id="label-control-plane-ui-0-87-0"></a>++### Control Plane UI 0.87.0++- Security patches and dependency upgrades.++<a id="label-data-plane-ui-0-20-0"></a>++### Data Plane UI 0.20.0++- Security patches and dependency upgrades.++<a id="label-runtime-operator-0-73-0"></a>++### Runtime Operator 0.73.0++- Security patches and dependency upgrades.++<a id="label-ingress-controller-2026-8-10-2"></a>++### Ingress Controller 2026.8.10-2++- Security patches and dependency upgrades.++<a id="label-spcs-data-plane-agent-1-44-0"></a>++### SPCS Data Plane Agent 1.44.0++- Security patches and dependency upgrades.++<a id="label-aws-data-plane-agent-1-60-0"></a>++### AWS Data Plane Agent 1.60.0++- A requested Deployment upgrade now triggers as soon as possible instead of+ waiting up to 10 minutes.+- Fixed issues that occasionally cause Deployment upgrades and teardowns to fail.+- Security patches and dependency upgrades.+ ## August 3, 2026
（この変更は要約対象外です。diff を参照してください）
変更内容: 本文を更新（43行追加・0行削除）
--- ahttps://docs.snowflake.com/en/user-guide/data-integration/openflow/connectors/oracle/about+++ bhttps://docs.snowflake.com/en/user-guide/data-integration/openflow/connectors/oracle/about@@ -528,4 +528,43 @@ - AWS Standard Single-tenant RDS for Oracle +<a id="label-oracle-standby-data-guard-support"></a>++## Data Guard and standby support++You can keep replication load off the primary Oracle database by connecting to a+Data Guard standby. Which topology you need depends on whether the standby is+physical or logical.++### Active Data Guard (physical standby)++Active Data Guard is a *physical* standby: a read-only copy of the primary. You+can't create an XStream outbound server on it, because XStream needs a writable+database for the metadata.++The connector still works with Active Data Guard. Use the physical standby for+snapshot load (`Oracle Connection URL`). For incremental (CDC) load, create the+XStream outbound server on a writable database (`XStream Out Server URL`): the+primary, or a separate+[downstream capture](https://docs.oracle.com/en/database/oracle/oracle-database/19/xstrm/xstream-out-concepts.html)+database if you want to keep CDC off the primary as well.++When you run the snapshot against Active Data Guard, set **Snapshot Fetching+Strategy** to `SEQUENTIAL_BY_PRIMARY_KEY`. The default `CONCURRENT_BY_ROWID`+strategy uses parallel tasks to split a table into ranges, which isn't allowed+on a read-only standby.++### Logical standby++A *logical* standby is open read-write and applies changes from the primary with+SQL Apply. The connector can use a logical standby for both snapshot and CDC, so+you don't need an additional database for XStream.++Before you create the outbound server or start the connector, set Database Guard+to `STANDBY`. Logical standbys default to `ALL`, which blocks the XStream client+with `ORA-16224: Database Guard is enabled`.++For setup steps, see+[](/user-guide/data-integration/openflow/connectors/oracle/setup-oracledb#label-oracle-standby-setup).+ ## Limitations @@ -536,4 +575,8 @@ - Oracle SaaS offerings such as Oracle Fusion Cloud Applications and NetSuite aren't supported.+- Active Data Guard physical standbys are read-only, so create the XStream+ outbound server on the primary, a logical standby, or a separate downstream+ capture database. For details, see+ [](#label-oracle-standby-data-guard-support). - The connector requires Openflow deployment version 0.55.0 or later for BYOC. - The Openflow runtime must be created after the required Openflow deployment
（この変更は要約対象外です。diff を参照してください）
変更内容: 本文を更新（22行追加・3行削除）
--- ahttps://docs.snowflake.com/en/user-guide/cortex-code/cortex-code-snowsight/cloud-agents+++ bhttps://docs.snowflake.com/en/user-guide/cortex-code/cortex-code-snowsight/cloud-agents@@ -31,10 +31,29 @@ - **No change to existing grants.** Cloud Agents does not modify any of your Snowflake RBAC settings or open up additional SQL access. - **Isolated execution.** Each session runs in its own container.-- **Outbound network is restricted, not disabled.** HTTPS egress is limited to platform-configured package registries and build tooling (for example PyPI, npm and Yarn, RubyGems, Rust/crates.io, the Go module proxy, Maven and Gradle, and similar hosts for NuGet, Pub, Hex, CPAN, CocoaPods, and Swift). Access to hosts outside that allowlist requires [external access integrations](/developer-guide/external-network-access/external-network-access-overview) that your Snowflake administrator configures. These integrations use the same primitives as UDFs and stored procedures.+- **Outbound network is restricted, not disabled.** HTTPS egress is limited to a fixed allowlist of package managers and related build tooling. For the full list, see [Allowed package manager egress](#label-cloud-agents-allowed-package-manager-egress). Access to hosts outside that allowlist requires [external access integrations](/developer-guide/external-network-access/external-network-access-overview) that your Snowflake administrator configures. These integrations use the same primitives as UDFs and stored procedures. - **Existing secrets model.** If external access is configured, credentials are managed via Snowflake Secrets. The container never has direct access to secret values.++<a id="label-cloud-agents-allowed-package-manager-egress"></a>++### Allowed package manager egress++By default, the Cloud Agents sandbox can reach the following package managers and related hosts over HTTPS so you can install dependencies and build tooling inside the container. Access to these hosts is read-only (GET and HEAD only), so package installs and downloads work, but publish and other write operations are blocked:++| Ecosystem | What's allowed |+| ------------------------- | --------------------------------------------------------------- |+| Python | PyPI |+| JavaScript | npm, Yarn, and the Node.js distribution server |+| Ruby | RubyGems and RVM |+| Rust | crates.io and rustup |+| Go | The Go module proxy, checksum database, and related Go hosts |+| JVM | Maven, Gradle, Kotlin, and Spring |+| Other language registries | Packagist, NuGet, Pub, Hex, CPAN, Hackage, CocoaPods, and Swift |+| OS packages | Debian, Ubuntu, and Alpine package mirrors |++Anything outside Snowflake and this allowlist, including arbitrary websites or private registries, isn't reachable unless your administrator configures an external access integration. ## Disable non-Snowflake egress (account administrators) -By default, Cloud Agents allows HTTPS egress to platform-configured package registries and build tooling. Account administrators can block all non-Snowflake egress from the sandbox for the entire account by setting the [COCO_CLOUD_AGENTS_NON_SNOWFLAKE_EGRESS_DISABLED](/sql-reference/parameters#label-coco-cloud-agents-non-snowflake-egress-disabled) account parameter to `TRUE`:+By default, Cloud Agents allows HTTPS egress to the package manager allowlist. Account administrators can block all non-Snowflake egress from the sandbox for the entire account by setting the [COCO_CLOUD_AGENTS_NON_SNOWFLAKE_EGRESS_DISABLED](/sql-reference/parameters#label-coco-cloud-agents-non-snowflake-egress-disabled) account parameter to `TRUE`: ```sql@@ -42,5 +61,5 @@ ``` -Only ACCOUNTADMIN can change this parameter. The default is `FALSE` (non-Snowflake egress to the platform allowlist is allowed). When set to `TRUE`, the sandbox cannot reach hosts outside Snowflake, including package registries.+Only ACCOUNTADMIN can change this parameter. The default is `FALSE` (non-Snowflake egress to the package manager allowlist is allowed). When set to `TRUE`, the sandbox cannot reach hosts outside Snowflake, including package registries. ## Current scope
（この変更は要約対象外です。diff を参照してください）
変更内容: 本文を更新（16行追加・5行削除）
--- ahttps://docs.snowflake.com/en/developer-guide/snowflake-app-runtime/app-yml+++ bhttps://docs.snowflake.com/en/developer-guide/snowflake-app-runtime/app-yml@@ -143,6 +143,6 @@ Presentation metadata surfaced alongside the deployed service. -- `icon`: Relative path to a `.png` or `.svg` file inside the package. The path- must not contain `..`.+- `icon`: Relative path to a `.png`, `.svg`, or `.webp` file inside the package.+ The path must not contain `..`, and it can't be an absolute path. - `label`: Short display label. - `description`: Longer human-readable description.@@ -165,6 +165,13 @@ variables inside the application container. Each entry has: -- `name`: Environment variable name.+- `name`: Environment variable name. Must match the POSIX pattern+ `[A-Z_][A-Z0-9_]*` (uppercase letters, digits, and underscores; must start+ with a letter or underscore). Duplicate names aren't allowed. - `value`: String value.++Some environment variable names are reserved for the platform and can't be set+here: any name starting with `SF_SNOWFS_`, plus `SNOWFLAKE_AUTH_MODE`,+`SNOWFLAKE_INSECURE_MODE`, `NODE_ENV`, `NEXT_PRIVATE_STANDALONE`, `PORT`, and+`HOSTNAME`. Setting a reserved name fails validation or is dropped at deploy. ```yaml@@ -193,5 +200,6 @@ to the application at runtime. Each entry has: -- `name`: Mount name used to locate the secret files inside the container.+- `name`: Mount name used to locate the secret files inside the container. Must+ match the POSIX pattern `[A-Z_][A-Z0-9_]*`. Duplicate names aren't allowed. - `secret`: Fully qualified name of an existing Snowflake secret (`<database>.<schema>.<secret_name>`).@@ -320,5 +328,8 @@ - Invalid `profile.icon` values (path traversal, absolute paths, or unsupported image formats).-- An unresolvable name in `secrets` or `external_access_integrations`.++Names in `secrets` and `external_access_integrations` are resolved when+Snowflake creates or upgrades the Application Service. An unresolvable name+fails that create or upgrade step, not the build phase. The error message includes parser details, for example:
（この変更は要約対象外です。diff を参照してください）
変更内容: 本文を更新（18行追加・0行削除）
--- ahttps://docs.snowflake.com/en/sql-reference/sql/show-delegated-authorizations+++ bhttps://docs.snowflake.com/en/sql-reference/sql/show-delegated-authorizations@@ -70,4 +70,22 @@ removed in a future [behavior change release](#label-behavior-change-bundles). +<div className="previewfeat sidebar">++<div className="sidebar-title">++%logo-snowflake-black% [Preview Feature](/release-notes/preview-features) — Open++</div>++Available to all accounts.++</div>++The output includes a `can_switch_roles` column. `TRUE` indicates that the delegated authorization permits in-session role switching.+For more information, see [](/user-guide/oauth-custom#label-oauth-custom-role-switching).++The output includes a `consented_roles` column listing the roles the user consented to for the delegated authorization. For more+information, see [](/user-guide/oauth-custom#label-oauth-custom-role-selection).+ ## Examples
（この変更は要約対象外です。diff を参照してください）
変更内容: 本文を更新（6行追加・6行削除）
--- ahttps://docs.snowflake.com/en/sql-reference/sql/create-application-service+++ bhttps://docs.snowflake.com/en/sql-reference/sql/create-application-service@@ -122,9 +122,9 @@ Requirements: -- The specified role must be granted to the owning role of the app.-- If omitted, Snowflake sets `EXECUTE_AS_ROLE` to the primary role of the- session that creates the app.--If the specified role is no longer granted to the owning role when the app starts,+- The specified role must be granted to the owning user of the app.+- If omitted, Snowflake sets `EXECUTE_AS_ROLE` to the creator's session+ primary role.++If the specified role is no longer granted to the owning user when the app starts, startup fails. @@ -203,5 +203,5 @@ <td> Required on the role specified in `EXECUTE_AS_ROLE`. The role must be- granted to the owning role of the app.+ granted to the owning user of the app. </td> </tr>
（この変更は要約対象外です。diff を参照してください）
変更内容: 本文を更新（6行追加・6行削除）
--- ahttps://docs.snowflake.com/en/developer-guide/snowflake-app-runtime/access-control+++ bhttps://docs.snowflake.com/en/developer-guide/snowflake-app-runtime/access-control@@ -50,20 +50,20 @@ - **Standard database**: the execution role is the role that owns the app,- which is the primary role of the session that created the app.+ which is the creator's session primary role. - **[Personal database](/user-guide/personal-databases) (PDB)**: the app is owned by the user who created it, not by a role, so you specify the execution role using `EXECUTE_AS_ROLE` in [](/sql-reference/sql/create-application-service). If you don't set it,- Snowflake uses the creator's primary session role. The execution role must be- granted to the owning role.+ Snowflake uses the creator's session primary role. The execution role must be+ granted to the owning user. Once the app is created, its execution role can't be changed. To use a different role, drop and recreate the app with the new role. -To set the execution role for a PDB app, grant the execution role to the owning role-and set `EXECUTE_AS_ROLE`:+To set the execution role for a PDB app, grant the execution role to the owning+user and set `EXECUTE_AS_ROLE`: ```sql-GRANT ROLE my_app_role TO ROLE <owning_role>;+GRANT ROLE my_app_role TO USER <owning_user>; CREATE APPLICATION SERVICE my_db.my_schema.my_app
（この変更は要約対象外です。diff を参照してください）
変更内容: 本文を更新（6行追加・0行削除）
--- ahttps://docs.snowflake.com/en/release-notes/bcr-bundles/un-bundled/unbundled-behavior-changes+++ bhttps://docs.snowflake.com/en/release-notes/bcr-bundles/un-bundled/unbundled-behavior-changes@@ -207,4 +207,10 @@ <tbody> <tr>+ <td>**August 20, 2026**</td>+ <td>Cortex Agents — Managed MCP Server</td>+ <td>[](/release-notes/bcr-bundles/un-bundled/bcr-2405)</td>+ <td>Managed MCP `tools/call` responses change from a single JSON body to an SSE stream. Spec-compliant clients are unaffected.</td>+ </tr>+ <tr> <td>**August 12, 2026**</td> <td>Cortex Model Updates for August</td>
（この変更は要約対象外です。diff を参照してください）
変更内容: 本文を更新（3行追加・3行削除）
--- ahttps://docs.snowflake.com/en/sql-reference/functions/data_quality_monitoring_expectation_status+++ bhttps://docs.snowflake.com/en/sql-reference/functions/data_quality_monitoring_expectation_status@@ -142,10 +142,10 @@ </tr> <tr>- <td>`arguments_ids`</td>+ <td>`argument_ids`</td> <td>ARRAY</td> <td>Array of the identifiers of the DMF arguments. Array elements are in the same order as the arguments.</td> </tr> <tr>- <td>`arguments_types`</td>+ <td>`argument_types`</td> <td>ARRAY</td> <td>@@ -158,5 +158,5 @@ </tr> <tr>- <td>`arguments_names`</td>+ <td>`argument_names`</td> <td>ARRAY</td> <td>
（この変更は要約対象外です。diff を参照してください）
変更内容: 本文を更新（3行追加・3行削除）
--- ahttps://docs.snowflake.com/en/sql-reference/local/data_quality_monitoring_expectation_status+++ bhttps://docs.snowflake.com/en/sql-reference/local/data_quality_monitoring_expectation_status@@ -95,10 +95,10 @@ </tr> <tr>- <td>`arguments_ids`</td>+ <td>`argument_ids`</td> <td>ARRAY</td> <td>Array of the identifiers of the DMF arguments. Array elements are in the same order as the arguments.</td> </tr> <tr>- <td>`arguments_types`</td>+ <td>`argument_types`</td> <td>ARRAY</td> <td>@@ -111,5 +111,5 @@ </tr> <tr>- <td>`arguments_names`</td>+ <td>`argument_names`</td> <td>ARRAY</td> <td>
（この変更は要約対象外です。diff を参照してください）
変更内容: 本文を更新（3行追加・3行削除）
--- ahttps://docs.snowflake.com/en/sql-reference/local/data_quality_monitoring_results+++ bhttps://docs.snowflake.com/en/sql-reference/local/data_quality_monitoring_results@@ -97,10 +97,10 @@ </tr> <tr>- <td>`arguments_ids`</td>+ <td>`argument_ids`</td> <td>ARRAY</td> <td>Array of the identifiers of the DMF arguments. Array elements are in the same order as the arguments.</td> </tr> <tr>- <td>`arguments_types`</td>+ <td>`argument_types`</td> <td>ARRAY</td> <td>@@ -113,5 +113,5 @@ </tr> <tr>- <td>`arguments_names`</td>+ <td>`argument_names`</td> <td>ARRAY</td> <td>
（この変更は要約対象外です。diff を参照してください）
変更内容: 本文を更新（3行追加・3行削除）
--- ahttps://docs.snowflake.com/en/sql-reference/functions/data_quality_monitoring_results+++ bhttps://docs.snowflake.com/en/sql-reference/functions/data_quality_monitoring_results@@ -139,10 +139,10 @@ </tr> <tr>- <td>`arguments_ids`</td>+ <td>`argument_ids`</td> <td>ARRAY</td> <td>Array of the identifiers of the DMF arguments. Array elements are in the same order as the arguments.</td> </tr> <tr>- <td>`arguments_types`</td>+ <td>`argument_types`</td> <td>ARRAY</td> <td>@@ -155,5 +155,5 @@ </tr> <tr>- <td>`arguments_names`</td>+ <td>`argument_names`</td> <td>ARRAY</td> <td>
（この変更は要約対象外です。diff を参照してください）
変更内容: 本文を更新（4行追加・1行削除）
--- ahttps://docs.snowflake.com/en/developer-guide/snowflake-app-runtime/privileges+++ bhttps://docs.snowflake.com/en/developer-guide/snowflake-app-runtime/privileges@@ -46,5 +46,8 @@ <tr> <td>USAGE</td>- <td>Access public endpoints exposed by the service.</td>+ <td>+ Appear in `SHOW APPLICATION SERVICES` results and access public+ endpoints exposed by the service.+ </td> </tr> <tr>
（この変更は要約対象外です。diff を参照してください）
変更内容: 本文を更新（2行追加・3行削除）
--- ahttps://docs.snowflake.com/en/user-guide/data-quality-ui-setup+++ bhttps://docs.snowflake.com/en/user-guide/data-quality-ui-setup@@ -1,7 +1,6 @@ # Use %sf-web-interface% to set up data quality checks -Available to all accounts that are Enterprise Edition (or higher).--To inquire about upgrading, please contact [Snowflake Support](https://docs.snowflake.com/user-guide/contacting-support).+Data Quality Monitoring requires Enterprise Edition. To inquire about upgrading, please contact+[Snowflake Support](https://docs.snowflake.com/user-guide/contacting-support). - [](/user-guide/data-quality-intro)
（この変更は要約対象外です。diff を参照してください）
変更内容: 本文を更新（2行追加・2行削除）
--- ahttps://docs.snowflake.com/en/release-notes/bcr-bundles/2026_06/bcr-2376+++ bhttps://docs.snowflake.com/en/release-notes/bcr-bundles/2026_06/bcr-2376@@ -39,7 +39,7 @@ | AWS | `eu-west-2` | Europe (London) | `AWS_EU` | | AWS | `eu-west-3` | EU (Paris) | `AWS_EU` |-| AWS | `ap-northeast-1` | Asia Pacific (Tokyo) | `AWS_APJ` |+| AWS | `ap-northeast-1` | Asia Pacific (Tokyo) | `AWS_JP` | | AWS | `ap-northeast-2` | Asia Pacific (Seoul) | `AWS_APJ` |-| AWS | `ap-northeast-3` | Asia Pacific (Osaka) | `AWS_APJ` |+| AWS | `ap-northeast-3` | Asia Pacific (Osaka) | `AWS_JP` | | AWS | `ap-south-1` | Asia Pacific (Mumbai) | `AWS_APJ` | | AWS | `ap-southeast-1` | Asia Pacific (Singapore) | `AWS_APJ` |
（この変更は要約対象外です。diff を参照してください）
変更内容: 本文を更新（2行追加・2行削除）
--- ahttps://docs.snowflake.com/en/collaboration/provider-listings-auto-fulfillment+++ bhttps://docs.snowflake.com/en/collaboration/provider-listings-auto-fulfillment@@ -213,6 +213,6 @@ - Snowflake compiles the listing auto-fulfillment refresh history and sends emails for failed listing refreshes daily. These messages are sent to the email address specified on the listing. - If the provider has a tag that includes a masking policy at the *account* level, auto-fulfillment doesn't take that masking policy into account when auto-fulfilling the data product. For auto-fulfillment, the scope of sharing is at the database, schema, and table level, but not at the account level.-- Auto-fulfillment enforces a 10TB limit on the size of the data product. For more information, refer to the- [](#label-laf-troubleshoot-database-too-large) troubleshooting topic. After assessing the cost implications, you can contact [Snowflake Support](https://docs.snowflake.com/user-guide/contacting-support) to increase the size limit.+- Auto-fulfillment enforces a default 10TB limit on the size of the data product. For more information, refer to the+ [](#label-laf-troubleshoot-database-too-large) troubleshooting topic. After assessing the cost implications, an account administrator can increase the size limit by setting the [](/sql-reference/parameters#label-listing-auto-fulfillment-initial-refresh-size-limit-in-tb) parameter. - If you use [Tri-Secret Secure](/user-guide/security-encryption-tss), you must contact
（この変更は要約対象外です。diff を参照してください）
変更内容: 3行以下の小規模な更新
--- ahttps://docs.snowflake.com/en/user-guide/interactive+++ bhttps://docs.snowflake.com/en/user-guide/interactive@@ -631,6 +631,9 @@ - `us-east-2` - AWS US East (Ohio) - `ca-central-1` - AWS Canada (Central)+- `af-south-1` - AWS Africa (Cape Town) - `ap-northeast-1` - AWS Asia Pacific (Tokyo)+- `ap-northeast-3` - AWS Asia Pacific (Osaka) - `ap-southeast-2` - AWS Asia Pacific (Sydney)+- `sa-east-1` - AWS South America (São Paulo) - `eu-central-1` - AWS EU (Frankfurt) - `eu-west-1` - AWS EU (Ireland)
（この変更は要約対象外です。diff を参照してください）
変更内容: 3行以下の小規模な更新
--- ahttps://docs.snowflake.com/en/user-guide/alerts-ui+++ bhttps://docs.snowflake.com/en/user-guide/alerts-ui@@ -258,6 +258,5 @@ | **DATA_QUALITY** | **Anomaly detection alert** | Monitors for [anomalies detected in data quality metrics](/user-guide/data-quality-anomaly), triggering an alert when unusual patterns or outliers are identified. | | | **Expectation violations alert** | Monitors for [data quality expectation violations](/user-guide/data-quality-expectations), triggering an alert when defined expectations are violated. |-| **OPENFLOW** | **High CPU alert** | Monitors Openflow runtime pods for sustained high CPU usage (>90%), which can indicate a potential performance degradation. |-| | **Connector backpressure (object count)** | Monitors the connector queue count to determine if it exceeds the backpressure threshold, which can indicate that the downstream system can't keep up with the data flow rate. |+| **OPENFLOW** | **Connector backpressure (object count)** | Monitors the connector queue count to determine if it exceeds the backpressure threshold, which can indicate that the downstream system can't keep up with the data flow rate. | | | **Connector backpressure (bytes)** | Monitors the connector queued bytes to determine if it exceeds the backpressure threshold, which can indicate that there is memory pressure in the data pipeline. | | | **High queued bytes alert** | Monitors the connector queue bytes to determine if it approaches the backpressure threshold (>80%), which can provide an early warning before backpressure occurs. |
（この変更は要約対象外です。diff を参照してください）
変更内容: 3行以下の小規模な更新
--- ahttps://docs.snowflake.com/en/sql-reference/account-usage/stage_storage_usage_details+++ bhttps://docs.snowflake.com/en/sql-reference/account-usage/stage_storage_usage_details@@ -187,5 +187,5 @@ - This view is available for accounts hosted on AWS and Azure. It is not supported on GCP. - The view contains historical data for the last 90 days. Data accumulation began on the date this feature became generally available.-- For AWS, the view displays information about named, table, and user stages. For Azure, the view only displays information for named stages.+- The view displays information about named, table, and user stages. - Snowflake makes a best effort to provide information for all days. However, there might be days for which stage usage details aren't available.
（この変更は要約対象外です。diff を参照してください）
変更内容: 3行以下の小規模な更新
--- ahttps://docs.snowflake.com/en/user-guide/security-access-control-privileges+++ bhttps://docs.snowflake.com/en/user-guide/security-access-control-privileges@@ -1356,5 +1356,5 @@ <tr> <td>USE_ANY_ROLE</td>- <td>Allows the External OAuth client or user to switch roles only if this privilege is granted to the client or user. Configure the External OAuth security integration to use the `EXTERNAL_OAUTH_ANY_ROLE_MODE` parameter using [CREATE SECURITY INTEGRATION](/sql-reference/sql/create-security-integration-oauth-external) or [ALTER SECURITY INTEGRATION](/sql-reference/sql/alter-security-integration-oauth-external).</td>+ <td>Allows an OAuth client or user to switch roles only if this privilege is granted to the client or user. Configure the External OAuth security integration to use the `EXTERNAL_OAUTH_ANY_ROLE_MODE` parameter using [CREATE SECURITY INTEGRATION](/sql-reference/sql/create-security-integration-oauth-external) or [ALTER SECURITY INTEGRATION](/sql-reference/sql/alter-security-integration-oauth-external), or configure the Snowflake OAuth security integration to use the `OAUTH_ANY_ROLE_MODE` parameter using [CREATE SECURITY INTEGRATION](/sql-reference/sql/create-security-integration-oauth-snowflake) or [ALTER SECURITY INTEGRATION](/sql-reference/sql/alter-security-integration-oauth-snowflake).</td> </tr> <tr>
（この変更は要約対象外です。diff を参照してください）
変更内容: 3行以下の小規模な更新
--- ahttps://docs.snowflake.com/en/sql-reference/sql/alter-account+++ bhttps://docs.snowflake.com/en/sql-reference/sql/alter-account@@ -86,4 +86,5 @@ EXTERNAL_OAUTH_ADD_PRIVILEGED_ROLES_TO_BLOCKED_LIST = TRUE | FALSE INITIAL_REPLICATION_SIZE_LIMIT_IN_TB = <num>+ LISTING_AUTO_FULFILLMENT_INITIAL_REFRESH_SIZE_LIMIT_IN_TB = <num> LISTING_AUTO_FULFILLMENT_REPLICATION_REFRESH_SCHEDULE = <schedule> LLM_INFERENCE_PARSE_DOCUMENT_PRESIGNED_URL_EXPIRY_SECONDS = <integer>
（この変更は要約対象外です。diff を参照してください）
変更内容: 3行以下の小規模な更新
--- ahttps://docs.snowflake.com/en/user-guide/data-integration/openflow/connectors/oracle/setup-connector+++ bhttps://docs.snowflake.com/en/user-guide/data-integration/openflow/connectors/oracle/setup-connector@@ -564,4 +564,5 @@ - **CONCURRENT_BY_ROWID** (default): Splits tables into chunks bound by ranges of physical row ids, and retrieves each chunk in parallel.+ This strategy isn't currently supported when the connector reads from a read-only database like Active Data Guard physical standby. - **SEQUENTIAL_BY_PRIMARY_KEY**: Uses fixed-size batches retrieved sequentially by the table's replication key (primary key, unique constraint, unique index, or logical key). Despite the name, this strategy uses whatever key the connector resolved for the table, not specifically the primary key.
（この変更は要約対象外です。diff を参照してください）
変更内容: 3行以下の小規模な更新
--- ahttps://docs.snowflake.com/en/sql-reference/service-types+++ bhttps://docs.snowflake.com/en/sql-reference/service-types@@ -57,4 +57,5 @@ | SERVERLESS_TASKS_FLEX | SERVERLESS TASKS FLEX | Flexible serverless compute for scheduled tasks. | Compute | Credits | | SNOWFLAKE_APP_RUNTIME | SNOWFLAKE APP RUNTIME | Compute used to build and host Snowflake Application Services on managed compute pools. | Compute | Credits |+| SNOWFLAKE_APP_RUNTIME_SERVERLESS | SNOWFLAKE APP RUNTIME SERVERLESS | Compute used to build and host Snowflake Application Services on serverless compute. | Compute | Credits | | SNOWFLAKE_COCO | SNOWFLAKE COCO | AI-powered coding assistant consumption billed via canonical Cortex Code events. | AI and Machine Learning | AI Credits | | SNOWFLAKE_COCO_CLI | SNOWFLAKE COCO: CLI | AI-powered coding assistant accessed via the command-line interface for local development. | AI and Machine Learning | AI Credits |
| セクション | S | A | B | C | 計 |
|---|---|---|---|---|---|
| User Guide | 0 | 1 | 2 | 3 | 6 |
| SQL Commands | 0 | 2 | 3 | 1 | 6 |
| Developer Guide | 0 | 3 | 3 | 0 | 6 |
| Loading & Unloading Data | 0 | 2 | 2 | 1 | 5 |
| Cortex Code | 1 | 0 | 3 | 0 | 4 |
| SQL General Reference | 0 | 0 | 3 | 1 | 4 |
| Streamlit in Snowflake | 0 | 2 | 0 | 0 | 2 |
| Collaboration & Marketplace | 0 | 1 | 1 | 0 | 2 |
| Release Notes | 0 | 0 | 2 | 0 | 2 |
| SQL Functions | 0 | 0 | 2 | 0 | 2 |
| Account Usage | 0 | 0 | 0 | 1 | 1 |