@thinkingai/ae-cli 6.0.18 → 6.0.22

package/skills/ae-analysis/references/freeze_dashboards.md

@@ -0,0 +1,29 @@
+# analysis +freeze_dashboards (Freeze / Unfreeze Dashboards)
+
+> **Prerequisite:** Follow the Global AE CLI Rules in [`../SKILL.md`](../SKILL.md).
+
+Domain: **Dashboard Management**
+
+## Use Cases
+- Freeze one or more dashboards to take their scheduled refresh jobs offline.
+- Unfreeze dashboards to bring scheduled jobs back online (for dashboards that have a schedule configured).
+
+## Commands
+```bash
+ae-cli analysis_audience +freeze_dashboards --project_id <project_id> --dashboard_ids '[123, 456]' --freeze true
+ae-cli analysis_audience +freeze_dashboards --project_id <project_id> --dashboard_ids '[123]' --freeze false
+ae-cli analysis_audience +freeze_dashboards --dry-run
+```
+
+## Parameters
+| Parameter | Required | Description |
+|---|---|---|
+| `--project_id` / `-p` | Yes | Project ID |
+| `--dashboard_ids` | Yes | JSON array of dashboard IDs to freeze or unfreeze |
+| `--freeze` | Yes | true to freeze, false to unfreeze |
+
+## Decision Rules
+- Use `+list_dashboards` to confirm dashboard IDs before freezing.
+
+## Recommended Chain
+- `+list_dashboards` -> `+freeze_dashboards`

package/skills/ae-analysis/references/get_table_columns.md

@@ -5,16 +5,16 @@
 Domain: **Model analysis**
 
 ## Use Cases
-- Query the field list of a project table. Returns all column names and types under the specified catalog, schema, and table so the table schema can be understood before SQL analysis.
+- Query the field list of a project table. Returns all column names and types for `tableRef` so the table schema can be understood before SQL analysis.
 - Table guide:
-- default.public.ta_event_1: catalog is default, schema is public, table is ta_event_1
-- public.ta_event_1: catalog is hive(default value), schema is public, table is ta_event_1
-- ta_event_1: catalog is hive(default value), schema is ta(default value), table is ta_event_1
-- Query the field list of a project table.
+- `hive.ta_dim.datatable_1`: catalog is hive, schema is ta_dim, table is datatable_1.
+- `ta_dim.datatable_1`: catalog defaults to hive, schema is ta_dim, table is datatable_1.
+- `ta_event_1`: table only; resolved against project available tables and fails if ambiguous.
+- The only common catalog is `hive`. Common hive schemas are `ta` (default analysis tables), `ta_dim` (dimension/datatable/exchange tables), `temp` (SQL temporary tables), and `ta_ext` (external datatable/API tables). Do not replace an explicit schema such as `ta_dim` with default `ta`.
 
 ## Command
 ```bash
-ae-cli analysis +get_table_columns --project_id <project_id> --catalog demo --schema demo --table demo
+ae-cli analysis +get_table_columns --project_id <project_id> --table_ref hive.ta_dim.datatable_1
 ae-cli analysis +get_table_columns --dry-run
 ```
 
@@ -22,17 +22,16 @@ ae-cli analysis +get_table_columns --dry-run
 | Parameter | Required | Description |
 |---|---|---|
 | `--project_id` / `-p` | Yes | Project ID |
-| `--catalog` | Yes | Catalog name, default hive |
-| `--schema` | Yes | Schema name, default ta |
-| `--table` | Yes | Table name |
+| `--table_ref` | Yes | Table reference. Supports `hive.schema.table`, `schema.table`, or `table`. If only `table` is provided, the backend resolves it against project available tables and fails on ambiguity. |
 
 ## Decision Rules
-- On the first run, start with only the required parameters (`--project_id`,`--catalog`,`--schema`), and add optional parameters after confirming the path works.
+- Prefer a fully qualified `--table_ref` when the schema is known, especially for `ta_dim`, `temp`, or `ta_ext` tables.
+- Use table-only `--table_ref <table>` only when you expect the table name to be unique in the project; if the backend returns ambiguity, retry with one of the returned fully qualified `tableRef` values.
 - For cross-project troubleshooting, first confirm whether `--project_id` matches the current permissions and target environment.
 
 ## Next Steps on Failure
-- If required parameters are missing, fall back to the smallest runnable command and fill them in first (focus on `--project_id`, `--catalog`, `--schema`).
-- If reading fails, first verify whether the object ID exists and belongs to the current project.
+- If required parameters are missing, fill in `--project_id` and `--table_ref` first.
+- If reading fails, first verify whether the `tableRef` exists, is unambiguous, and belongs to the current project permissions.
 
 ## Recommended chaining
 - +get_table_columns

package/skills/ae-analysis/references/list_cluster_members.md

@@ -28,17 +28,17 @@ ae-cli analysis_audience +list_cluster_members --dry-run
 | `--fields` | No | Optional return field list (JSON array). Invalid fields will fail with `INVALID_FIELDS`. |
 | `--limit` | No | Optional page size. Default: 20, max: 50. |
 | `--offset` | No | Optional page offset. Default: 0. |
-| `--request_id` | No | Optional unique request ID used for tracking and cancellation. If provided, it must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`. For long-running or cancelable queries, provide this before starting the query so it can be cancelled later with `+cancel_query --request_id <same value>`, even if the caller stops waiting before the tool returns. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running. The auto-generated requestId is not available when the HTTP request fails before a response, so preset `requestId` is required for proactive cleanup. Generated automatically if omitted. The response `metadata.requestId` can also be passed to `cancel_query` when the query is no longer needed. |
+| `--request_id` | Yes | Required unique request ID used for tracking and cancellation. Generate it before starting the query. It must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`. Provide this before starting the query so it can be cancelled later with `+cancel_query --request_id <same value>`, even if the caller stops waiting before the tool returns. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running. `requestId` is not auto-generated for MCP query tools because the caller must know it before the response for proactive cleanup. If omitted or blank, the backend returns `REQUEST_ID_REQUIRED`; invalid format returns `INVALID_REQUEST_ID`. The response `metadata.requestId` echoes the supplied requestId and can also be passed to `cancel_query(requestId)` when the query is no longer needed. |
 | `--timeout_minutes` | No | Query timeout in minutes. If omitted, 30 minutes is used. |
 
 ## Decision Rules
-- For the first execution, it is recommended to pass only the required parameters (`--project_id`, `--cluster_name`) and add optional parameters after confirming the path works.
+- For the first execution, include all required parameters (`--project_id`, `--cluster_name`, `--request_id`) and add optional parameters after confirming the path works.
 - `--property_names` must be passed as a JSON array (e.g. `--property_names '["#user_id"]'`).
-- For long-running or cancelable member queries, supply your own `--request_id` before starting so `analysis +cancel_query` can cancel by the same ID if the caller or user stops waiting. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running; call `+cancel_query --request_id <same value>` with the preset ID. The auto-generated requestId is not available when the HTTP request fails before a response. The value must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`.
+- Generate and pass `--request_id` before starting any cancelable query; it is required. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running; call `+cancel_query --request_id <same value>` with the preset ID. The value must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`. Omitted or blank IDs return `REQUEST_ID_REQUIRED`; invalid format returns `INVALID_REQUEST_ID`.
 - When troubleshooting across projects, first confirm whether `--project_id` matches the current permissions and target environment.
 
 ## Next Step on Failure
-- If required parameters are missing, fall back to the smallest runnable command and fill the gap (focus on `--project_id`, `--cluster_name`).
+- If required parameters are missing, fall back to the smallest runnable command and fill the gap (focus on `--project_id`, `--cluster_name`, `--request_id`).
 - If the result is empty, first confirm the project ID/keyword, then try broadening the filter conditions.
 
 ## Recommended Chain

package/skills/ae-analysis/references/list_clusters.md

@@ -19,7 +19,7 @@ Domain: **Cluster Management**
 ```bash
 ae-cli analysis_audience +list_clusters --project_id <project_id>
 ae-cli analysis_audience +list_clusters --project_id <project_id> --query demo
-ae-cli analysis_audience +list_clusters --project_id <project_id> --fields '["id","clusterName","usersNum"]' --limit 20 --offset 0
+ae-cli analysis_audience +list_clusters --project_id <project_id> --fields '["id","clusterName","displayName","remarks","clusterType","progress","usersNum"]' --limit 20 --offset 0
 ae-cli analysis_audience +list_clusters --dry-run
 ```
 
@@ -27,12 +27,14 @@ ae-cli analysis_audience +list_clusters --dry-run
 | Parameter | Required | Description |
 |---|---|---|
 | `--project_id` / `-p` | Yes | Project ID |
-| `--query` / `-q` | No | Optional keyword filter. Performs fuzzy matching against cluster names, display names, and remarks; if omitted, all clusters are returned. |
-| `--fields` | No | Optional return field list (JSON array). Invalid fields will fail with `INVALID_FIELDS`. |
+| `--query` / `-q` | No | Optional keyword filter. Performs fuzzy matching against clusterName, displayName, and remarks; if omitted, all clusters are returned. |
+| `--fields` | No | Optional return field list (JSON array). Supported fields: `id`, `clusterName`, `displayName`, `clusterType`, `progress`, `usersNum`, `refreshStatus`, `remarks`, `authenticationStatus`. Default fields when omitted: `id`, `clusterName`, `displayName`, `remarks`, `clusterType`, `progress`, `usersNum`, `authenticationStatus`. Invalid fields will fail with `INVALID_FIELDS`. |
 | `--limit` | No | Optional page size. Default: 20, max: 50. |
 | `--offset` | No | Optional page offset. Default: 0. |
+| `--authenticated_only` | No | When true, return only authenticated clusters. |
 
 ## Decision Rules
+- Use `--authenticated_only true` only when the user explicitly asks for authenticated assets. `authenticationStatus` is `1` for authenticated and `0` for unauthenticated.
 - For the first execution, it is recommended to pass only the required parameters (`--project_id`) and add optional parameters after confirming the path works.
 - When troubleshooting across projects, first confirm whether `--project_id` matches the current permissions and target environment.
 

package/skills/ae-analysis/references/list_dashboards.md

@@ -18,7 +18,7 @@ Domain: **Dashboard management**
 ```bash
 ae-cli analysis +list_dashboards --project_id <project_id>
 ae-cli analysis +list_dashboards --project_id <project_id> --query demo
-ae-cli analysis +list_dashboards --project_id <project_id> --fields '["dashboardId","dashboardName"]' --limit 20 --offset 0
+ae-cli analysis +list_dashboards --project_id <project_id> --fields '["dashboardId","dashboardName","remark"]' --limit 20 --offset 0
 ae-cli analysis +list_dashboards --dry-run
 ```
 
@@ -26,8 +26,8 @@ ae-cli analysis +list_dashboards --dry-run
 | Parameter | Required | Description |
 |---|---|---|
 | `--project_id` / `-p` | Yes | Project ID |
-| `--query` / `-q` | No | Optional keyword filter. Performs fuzzy matching against dashboard names and AI remarks; if omitted, all accessible dashboards are returned. |
-| `--fields` / `-f` | No | Optional fields to return (JSON array). Supported fields: `dashboardId`, `dashboardName`, `aiRemark`. |
+| `--query` / `-q` | No | Optional keyword filter. Performs fuzzy matching against dashboard names and remarks; if omitted, all accessible dashboards are returned. |
+| `--fields` / `-f` | No | Optional fields to return (JSON array). Supported fields: `dashboardId`, `dashboardName`, `remark`. Default fields when omitted: `dashboardId`, `dashboardName`, `remark`. Dashboard metadata exposes `remark` but no separate desc field. |
 | `--limit` / `-l` | No | Optional page size. Default: 20, maximum: 10000. |
 | `--offset` / `-o` | No | Optional page offset. Default: 0. |
 

package/skills/ae-analysis/references/list_entities.md

@@ -23,7 +23,7 @@ ae-cli analysis_meta +list_entities --dry-run
 | `--project_id` / `-p` | Yes | Project ID |
 | `--event_name` | No | Optional event name filter |
 | `--query` / `-q` | No | Optional keyword filter. Fuzzy match is applied to entityName, columnName, and columnDesc; if omitted, all accessible dashboards are returned. |
-| `--fields` / `-f` | No | Optional fields to return (JSON array). Supported fields: `entityId`, `entityName`, `columnName`, `columnDesc`, `selectType`. |
+| `--fields` / `-f` | No | Optional fields to return (JSON array). Supported fields: `entityId`, `entityName`, `columnName`, `columnDesc`, `selectType`, `tableType`, `entityType`. Default fields when omitted: `entityId`, `entityName`, `columnName`, `columnDesc`, `selectType`. Entity metadata exposes `columnDesc` as the description field; no remark field is available in this list response. |
 | `--limit` / `-l` | No | Optional page size. Default: 20, maximum: 10000. |
 | `--offset` / `-o` | No | Optional page offset. Default: 0. |
 

package/skills/ae-analysis/references/list_events.md

@@ -18,7 +18,7 @@ Domain: **Metadata Query**
 ```bash
 ae-cli analysis_meta +list_events --project_id <project_id>
 ae-cli analysis_meta +list_events --project_id <project_id> --query demo
-ae-cli analysis_meta +list_events --project_id <project_id> --query demo --fields '["eventName", "eventDesc"]' --limit 20 --offset 0
+ae-cli analysis_meta +list_events --project_id <project_id> --query demo --fields '["eventId", "eventName", "eventDesc", "remark"]' --limit 20 --offset 0
 ae-cli analysis_meta +list_events --dry-run
 ```
 
@@ -26,11 +26,14 @@ ae-cli analysis_meta +list_events --dry-run
 | Parameter | Required | Description |
 |---|---|---|
 | `--project_id` / `-p` | Yes | Project ID |
-| `--query` / `-q` | No | Optional keyword filter. Performs fuzzy matching against eventName, eventDesc, and aiRemark; if omitted, all events are returned. |
-| `--fields` / `-f` | No | Optional fields to return (JSON array). Supported fields: `eventName`, `eventDesc`, `aiRemark`, `eventTag`, `remark`. |
-| `--limit` / `-l` | No | Optional page size. Default: 20, maximum: 10000. |
+| `--query` / `-q` | No | Optional keyword filter. Performs fuzzy matching against eventName, eventDesc, and remark; if omitted, all events are returned. |
+| `--fields` / `-f` | No | Optional fields to return (JSON array). Supported fields: `eventId`, `eventName`, `eventDesc`, `remark`, `eventTag`, `authenticationStatus`. Default fields when omitted: `eventId`, `eventName`, `eventDesc`, `remark`, `authenticationStatus`. |
+| `--limit` / `-l` | No | Optional page size. Default: 20, maximum: 50. |
 | `--offset` / `-o` | No | Optional page offset. Default: 0. |
+| `--authenticated_only` | No | When true, return only authenticated events. |
+
 ## Decision Rules
+- Use `--authenticated_only true` only when the user explicitly asks for authenticated assets. `authenticationStatus` is `1` for authenticated and `0` for unauthenticated.
 - For the first run, pass only the required parameter (`--project_id`) to confirm the path works, then add optional parameters.
 - For builder-supported ad-hoc analysis, pass the user's event wording to the builder instead of pre-querying event metadata.
 - For cross-project troubleshooting, first confirm whether `--project_id` matches the current permissions and target environment.

package/skills/ae-analysis/references/list_metrics.md

@@ -20,7 +20,7 @@ Domain: **Metadata Query**
 ```bash
 ae-cli analysis_meta +list_metrics --project_id <project_id>
 ae-cli analysis_meta +list_metrics --project_id <project_id> --query demo
-ae-cli analysis_meta +list_metrics --project_id <project_id> --query demo --fields '["metricId","metricName"]' --limit 10 --offset 0
+ae-cli analysis_meta +list_metrics --project_id <project_id> --query demo --fields '["metricId","metricName","metricDesc","metricRemark","metricMode"]' --limit 10 --offset 0
 ae-cli analysis_meta +list_metrics --dry-run
 ```
 
@@ -29,11 +29,13 @@ ae-cli analysis_meta +list_metrics --dry-run
 |---|---|---|
 | `--project_id` / `-p` | Yes | Project ID |
 | `--query` / `-q` | No | Optional keyword filter. Fuzzy match on metricName, metricDesc, metricRemark. |
-| `--fields` | No | Optional fields to return. Supported: metricId, metricName, metricDesc, metricRemark, metricMode, openId, creator, creatorLoginName, updateOpenId, updateCreator, updateLoginName, createTime, updateTime. Invalid fields cause INVALID_FIELDS error. |
-| `--limit` | No | Optional limit. Default: 20, maximum: 10000. |
+| `--fields` | No | Optional fields to return. Supported: metricId, metricName, metricDesc, metricRemark, metricMode, authenticationStatus, openId, creator, creatorLoginName, updateOpenId, updateCreator, updateLoginName, createTime, updateTime. Default fields when omitted: metricId, metricName, metricDesc, metricRemark, metricMode, authenticationStatus. Invalid fields cause INVALID_FIELDS error. |
+| `--limit` | No | Optional limit. Default: 20, maximum: 50. |
 | `--offset` | No | Optional offset. Default: 0. |
+| `--authenticated_only` | No | When true, return only authenticated metrics. |
 
 ## Decision Rules
+- Use `--authenticated_only true` only when the user explicitly asks for authenticated assets. `authenticationStatus` is `1` for authenticated and `0` for unauthenticated.
 - First run should only pass the required parameter (`--project_id`), and add optional parameters only after the path is confirmed to work.
 - For pagination, use `--limit` and `--offset` together. Default limit is 20.
 - Use `--fields` to select specific columns for lighter response payloads.

package/skills/ae-analysis/references/list_properties.md

@@ -16,7 +16,7 @@ Domain: **Metadata Query**
 ```bash
 ae-cli analysis_meta +list_properties --project_id <project_id>
 ae-cli analysis_meta +list_properties --project_id <project_id> --scope event --event_name purchase --query demo
-ae-cli analysis_meta +list_properties --project_id <project_id> --query demo --fields '["propName", "propDesc"]' --limit 20 --offset 0
+ae-cli analysis_meta +list_properties --project_id <project_id> --query demo --fields '["propId", "propName", "propDesc", "remark", "selectType", "tableType"]' --limit 20 --offset 0
 ae-cli analysis_meta +list_properties --dry-run
 ```
 
@@ -26,13 +26,15 @@ ae-cli analysis_meta +list_properties --dry-run
 | `--project_id` / `-p` | Yes | Project ID |
 | `--scope` | No | Optional property scope. event means event properties, user means user properties. If omitted, all scopes are queried. |
 | `--event_name` | No | Optional event name. If provided, only event properties available for that event are returned. |
-| `--query` / `-q` | No | Optional keyword filter. Fuzzy match is applied to propName, propDesc, and aiRemark; if omitted, all matched properties are returned. |
-| `--fields` / `-f` | No | Optional fields to return (JSON array). Supported fields: `propName`, `propDesc`, `aiRemark`, `selectType`, `tableType`, `subTableType`. |
-| `--limit` / `-l` | No | Optional page size. Default: 20, maximum: 10000. |
+| `--query` / `-q` | No | Optional keyword filter. Fuzzy match is applied to propName, propDesc, and remark; if omitted, all matched properties are returned. |
+| `--fields` / `-f` | No | Optional fields to return (JSON array). Supported fields: `propId`, `propName`, `propDesc`, `remark`, `selectType`, `tableType`, `subTableType`, `authenticationStatus`. Default fields when omitted: `propId`, `propName`, `propDesc`, `remark`, `selectType`, `tableType`, `authenticationStatus`. |
+| `--limit` / `-l` | No | Optional page size. Default: 20, maximum: 50. |
 | `--offset` / `-o` | No | Optional page offset. Default: 0. |
+| `--authenticated_only` | No | When true, return only authenticated properties. |
 
 
 ## Decision Rules
+- Use `--authenticated_only true` only when the user explicitly asks for authenticated assets. `authenticationStatus` is `1` for authenticated and `0` for unauthenticated.
 - For the first run, pass only the required parameter (`--project_id`) to confirm the path works, then add optional parameters.
 - For builder-supported ad-hoc analysis, pass the user's property wording to the builder instead of pre-querying property metadata.
 - For cross-project troubleshooting, first confirm whether `--project_id` matches the current permissions and target environment.

package/skills/ae-analysis/references/list_reports.md

@@ -18,7 +18,7 @@ Domain: **Report management**
 ```bash
 ae-cli analysis +list_reports --project_id <project_id>
 ae-cli analysis +list_reports --project_id <project_id> --query demo
-ae-cli analysis +list_reports --project_id <project_id> --fields '["reportId","reportName","reportModel"]' --limit 20 --offset 0
+ae-cli analysis +list_reports --project_id <project_id> --fields '["reportId","reportName","reportDesc","remark","reportModel"]' --limit 20 --offset 0
 ae-cli analysis +list_reports --dry-run
 ```
 
@@ -26,8 +26,8 @@ ae-cli analysis +list_reports --dry-run
 | Parameter | Required | Description |
 |---|---|---|
 | `--project_id` / `-p` | Yes | Project ID |
-| `--query` / `-q` | No | Optional keyword filter. Performs fuzzy matching against report names, descriptions, and AI remarks; if omitted, all accessible reports are returned. |
-| `--fields` / `-f` | No | Optional fields to return (JSON array). Supported fields: `reportId`, `reportName`, `reportDesc`, `reportModel`, `aiRemark`. |
+| `--query` / `-q` | No | Optional keyword filter. Performs fuzzy matching against reportName, reportDesc, and remark; if omitted, all accessible reports are returned. |
+| `--fields` / `-f` | No | Optional fields to return (JSON array). Supported fields: `reportId`, `reportName`, `reportDesc`, `remark`, `reportModel`. Default fields when omitted: `reportId`, `reportName`, `reportDesc`, `remark`, `reportModel`. |
 | `--limit` / `-l` | No | Optional page size. Default: 20, maximum: 10000. |
 | `--offset` / `-o` | No | Optional page offset. Default: 0. |
 

package/skills/ae-analysis/references/list_spaces.md

@@ -0,0 +1,27 @@
+# analysis +list_spaces (List Spaces)
+
+> **Prerequisite:** Follow the Global AE CLI Rules in [`../SKILL.md`](../SKILL.md).
+
+Domain: **Dashboard Management**
+
+## Use Cases
+- List all spaces and folders accessible to the current user as a recursive tree.
+- Returns two parts: (1) project spaces with spaceId, name, spaceDesc, and children (folders/dashboards); (2) personal-space folders (My Space / Shared with me / Ungrouped) with their children.
+- Use to resolve `spaceId` for `+create_dashboard` / `+copy_dashboard`, and `toSpaceId` / `toFolderId` / `fromSpaceId` / `fromFolderId` for `+move_dashboard` and `+copy_dashboard`.
+
+## Commands
+```bash
+ae-cli analysis_audience +list_spaces --project_id <project_id>
+ae-cli analysis_audience +list_spaces --dry-run
+```
+
+## Parameters
+| Parameter | Required | Description |
+|---|---|---|
+| `--project_id` / `-p` | Yes | Project ID |
+
+## Decision Rules
+- Call this before `+copy_dashboard` or `+move_dashboard` whenever the user specifies a target space or folder by name.
+
+## Recommended Chain
+- `+list_spaces` -> `+copy_dashboard` / `+move_dashboard` / `+create_dashboard`

package/skills/ae-analysis/references/list_tag_members.md

@@ -29,17 +29,17 @@ ae-cli analysis_audience +list_tag_members --dry-run
 | `--fields` | No | Optional return field list (JSON array). Invalid fields will fail with `INVALID_FIELDS`. |
 | `--limit` | No | Optional page size. Default: 20, max: 50. |
 | `--offset` | No | Optional page offset. Default: 0. |
-| `--request_id` | No | Optional unique request ID used for tracking and cancellation. If provided, it must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`. For long-running or cancelable queries, provide this before starting the query so it can be cancelled later with `+cancel_query --request_id <same value>`, even if the caller stops waiting before the tool returns. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running. The auto-generated requestId is not available when the HTTP request fails before a response, so preset `requestId` is required for proactive cleanup. Generated automatically if omitted. The response `metadata.requestId` can also be passed to `cancel_query` when the query is no longer needed. |
+| `--request_id` | Yes | Required unique request ID used for tracking and cancellation. Generate it before starting the query. It must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`. Provide this before starting the query so it can be cancelled later with `+cancel_query --request_id <same value>`, even if the caller stops waiting before the tool returns. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running. `requestId` is not auto-generated for MCP query tools because the caller must know it before the response for proactive cleanup. If omitted or blank, the backend returns `REQUEST_ID_REQUIRED`; invalid format returns `INVALID_REQUEST_ID`. The response `metadata.requestId` echoes the supplied requestId and can also be passed to `cancel_query(requestId)` when the query is no longer needed. |
 | `--timeout_minutes` | No | Query timeout in minutes. If omitted, 30 minutes is used. |
 
 ## Decision Rules
-- For the first execution, it is recommended to pass only the required parameters (`--project_id`, `--tag_name`) and add optional parameters after confirming the path works.
+- For the first execution, include all required parameters (`--project_id`, `--tag_name`, `--request_id`) and add optional parameters after confirming the path works.
 - `--property_names` must be passed as a JSON array (e.g. `--property_names '["#user_id"]'`).
-- For long-running or cancelable member queries, supply your own `--request_id` before starting so `analysis +cancel_query` can cancel by the same ID if the caller or user stops waiting. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running; call `+cancel_query --request_id <same value>` with the preset ID. The auto-generated requestId is not available when the HTTP request fails before a response. The value must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`.
+- Generate and pass `--request_id` before starting any cancelable query; it is required. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running; call `+cancel_query --request_id <same value>` with the preset ID. The value must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`. Omitted or blank IDs return `REQUEST_ID_REQUIRED`; invalid format returns `INVALID_REQUEST_ID`.
 - When troubleshooting across projects, first confirm whether `--project_id` matches the current permissions and target environment.
 
 ## Next Step on Failure
-- If required parameters are missing, fall back to the smallest runnable command and fill the gap (focus on `--project_id`, `--tag_name`).
+- If required parameters are missing, fall back to the smallest runnable command and fill the gap (focus on `--project_id`, `--tag_name`, `--request_id`).
 - If the result is empty, first confirm the project ID/keyword, then try broadening the filter conditions.
 
 ## Recommended Chain

package/skills/ae-analysis/references/list_tags.md

@@ -17,7 +17,7 @@ Domain: **Tag Management**
 ```bash
 ae-cli analysis_audience +list_tags --project_id <project_id>
 ae-cli analysis_audience +list_tags --project_id <project_id> --query demo
-ae-cli analysis_audience +list_tags --project_id <project_id> --fields '["id","clusterName","usersNum"]' --limit 20 --offset 0
+ae-cli analysis_audience +list_tags --project_id <project_id> --fields '["id","clusterName","displayName","remarks","clusterType","subConditionTabType","progress","usersNum"]' --limit 20 --offset 0
 ae-cli analysis_audience +list_tags --dry-run
 ```
 
@@ -26,11 +26,13 @@ ae-cli analysis_audience +list_tags --dry-run
 |---|---|---|
 | `--project_id` / `-p` | Yes | Project ID |
 | `--query` / `-q` | No | Optional keyword filter. Performs fuzzy matching against tag names, display names, and remarks; if omitted, all tags are returned. |
-| `--fields` | No | Optional return field list (JSON array). Invalid fields will fail with `INVALID_FIELDS`. |
+| `--fields` | No | Optional return field list (JSON array). Supported fields: `id`, `clusterName`, `displayName`, `clusterType`, `subConditionTabType`, `progress`, `usersNum`, `remarks`, `authenticationStatus`. Default fields when omitted: `id`, `clusterName`, `displayName`, `remarks`, `clusterType`, `subConditionTabType`, `progress`, `usersNum`, `authenticationStatus`. Invalid fields will fail with `INVALID_FIELDS`. |
 | `--limit` | No | Optional page size. Default: 20, max: 50. |
 | `--offset` | No | Optional page offset. Default: 0. |
+| `--authenticated_only` | No | When true, return only authenticated tags. |
 
 ## Decision Rules
+- Use `--authenticated_only true` only when the user explicitly asks for authenticated assets. `authenticationStatus` is `1` for authenticated and `0` for unauthenticated.
 - For the first execution, it is recommended to pass only the required parameters (`--project_id`) and add optional parameters after confirming the path works.
 - When troubleshooting across projects, first confirm whether `--project_id` matches the current permissions and target environment.
 

package/skills/ae-analysis/references/move_dashboard.md

@@ -0,0 +1,33 @@
+# analysis +move_dashboard (Move Dashboard)
+
+> **Prerequisite:** Follow the Global AE CLI Rules in [`../SKILL.md`](../SKILL.md).
+
+Domain: **Dashboard Management**
+
+## Use Cases
+- Move a dashboard to a different space or folder.
+- Provide `--to_space_id` to move to a space root, or both `--to_space_id` and `--to_folder_id` to move into a folder within that space.
+- `--from_space_id` and `--from_folder_id` are optional; provide them when the dashboard exists in multiple spaces to disambiguate the source path.
+
+## Commands
+```bash
+ae-cli analysis_audience +move_dashboard --project_id <project_id> --dashboard_id <dashboard_id> --to_space_id <space_id>
+ae-cli analysis_audience +move_dashboard --project_id <project_id> --dashboard_id <dashboard_id> --to_space_id <space_id> --to_folder_id <folder_id>
+ae-cli analysis_audience +move_dashboard --dry-run
+```
+
+## Parameters
+| Parameter | Required | Description |
+|---|---|---|
+| `--project_id` / `-p` | Yes | Project ID |
+| `--dashboard_id` | Yes | Dashboard ID to move |
+| `--to_space_id` | Yes | Target space ID |
+| `--to_folder_id` | No | Target folder ID within the target space. Omit to place at the space root. |
+| `--from_space_id` | No | Source space ID. Omit if unknown. |
+| `--from_folder_id` | No | Source folder ID. Omit if unknown. |
+
+## Decision Rules
+- Call `+list_spaces` first to resolve `to_space_id` and `to_folder_id`.
+
+## Recommended Chain
+- `+list_spaces` -> `+move_dashboard`

package/skills/ae-analysis/references/query_adhoc.md

@@ -89,7 +89,7 @@ ae-cli analysis +query_adhoc --project_id <project_id> --model_type <model_type>
 | `--project_id` / `-p` | Yes | Project ID used to identify the analysis project |
 | `--model_type` | Yes | Model type. Supported values: event, retention, funnel, distribution, attribution, heat_map, interval, path, rank_list, prop_analysis, sql. |
 | `--qp` | Yes | Query parameter JSON. For `event`/`retention`/`funnel`/`prop_analysis`, pass QP returned by the matching builder tool. For non-builder models, construct QP through the legacy schema/metadata path. |
-| `--request_id` | No | Optional unique request ID used for tracking and cancellation. If provided, it must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`. For long-running or cancelable queries, provide this before starting the query so it can be cancelled later with `+cancel_query --request_id <same value>`, even if the caller stops waiting before the tool returns. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running. The auto-generated requestId is not available when the HTTP request fails before a response, so preset `requestId` is required for proactive cleanup. Generated automatically if omitted. The response `metadata.requestId` can also be passed to `cancel_query` when the query is no longer needed. |
+| `--request_id` | Yes | Required unique request ID used for tracking and cancellation. Generate it before starting the query. It must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`. Provide this before starting the query so it can be cancelled later with `+cancel_query --request_id <same value>`, even if the caller stops waiting before the tool returns. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running. `requestId` is not auto-generated for MCP query tools because the caller must know it before the response for proactive cleanup. If omitted or blank, the backend returns `REQUEST_ID_REQUIRED`; invalid format returns `INVALID_REQUEST_ID`. The response `metadata.requestId` echoes the supplied requestId and can also be passed to `cancel_query(requestId)` when the query is no longer needed. |
 | `--use_cache` | No | Whether to use result cache. Default: true |
 | `--zone_offset` | No | Time zone offset in hours. For example, UTC+8 is 8 and UTC-5 is -5 |
 | `--is_sort_by_columns` | No | Whether to sort query results by columns. Default: false |
@@ -100,7 +100,7 @@ ae-cli analysis +query_adhoc --project_id <project_id> --model_type <model_type>
 | `--timeout_minutes` | No | Query timeout in minutes. If the query exceeds this time, it will be cancelled automatically. |
 
 ## Decision Rules
-- On the first run, start with only the required parameters (`--project_id`,`--model_type`,`--qp`) only when the query is clearly small. For any query that may exceed the CLI/MCP HTTP timeout, include a preset `--request_id` from the first run.
+- On the first run, include all required parameters (`--project_id`,`--model_type`,`--qp`,`--request_id`). Generate `--request_id` before starting the query so cancellation can use the same value if the caller stops waiting.
 - Do not call this command with placeholder QP such as `{}`. For builder-supported models, wait for builder `status=generated`; for non-builder models, build QP from verified schema/metadata first.
 - For builder-supported models, do not run metadata/schema lookup to "help" the builder. The builder is the metadata resolver.
 - For pagination, use `--limit` and `--offset` together. Default limit is 1000, maximum 100000.

package/skills/ae-analysis/references/query_bi_panel_data.md

@@ -28,7 +28,7 @@ ae-cli analysis +query_bi_panel_data --project_id <project_id> --panel_id <panel
 | `--panel_id` | Yes | BI panel ID returned by `+list_bi_panels` |
 | `--page_key` | Yes | Page key returned by `+get_bi_panel_detail` |
 | `--result_type` | Yes | `charts` or `summary` |
-| `--request_id` | No | Optional unique request ID used for tracking and cancellation. If provided, it must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`. For long-running or cancelable queries, provide this before starting the query so it can be cancelled later with `+cancel_query --request_id <same value>`, even if the caller stops waiting before the tool returns. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running. The auto-generated requestId is not available when the HTTP request fails before a response, so preset `requestId` is required for proactive cleanup. Generated automatically if omitted. The response `metadata.requestId` can also be passed to `cancel_query` when the query is no longer needed. |
+| `--request_id` | Yes | Required unique request ID used for tracking and cancellation. Generate it before starting the query. It must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`. Provide this before starting the query so it can be cancelled later with `+cancel_query --request_id <same value>`, even if the caller stops waiting before the tool returns. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running. `requestId` is not auto-generated for MCP query tools because the caller must know it before the response for proactive cleanup. If omitted or blank, the backend returns `REQUEST_ID_REQUIRED`; invalid format returns `INVALID_REQUEST_ID`. The response `metadata.requestId` echoes the supplied requestId and can also be passed to `cancel_query(requestId)` when the query is no longer needed. |
 | `--chart_ids` | No | JSON array of chart IDs when `result_type=charts`. If omitted, all queryable charts on the page are queried. |
 | `--parameter_controls` | No | JSON array of dashboard-level parameter control overrides. Each item has `controlId` and one scalar `value`. |
 | `--permission_controls` | No | JSON array of dashboard-level permission control values. Each item has `controlId` and `value`; `value` may be one scalar or an array. |
@@ -52,7 +52,7 @@ ae-cli analysis +query_bi_panel_data --project_id <project_id> --panel_id <panel
 - For charts, omit `--chart_ids` on the first run to query all queryable charts on the page.
 - Explicitly querying a non-queryable filter component returns a chart-level unsupported result.
 - Use `--columns`, `--row_limit`, and `--row_offset` to reduce response size.
-- For long-running or cancelable queries, supply your own `--request_id` before starting so `+cancel_query` can cancel by the same ID if the caller or user stops waiting. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running; call `+cancel_query --request_id <same value>` with the preset ID. The auto-generated requestId is not available when the HTTP request fails before a response. The value must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`.
+- Generate and pass `--request_id` before starting any cancelable query; it is required. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running; call `+cancel_query --request_id <same value>` with the preset ID. The value must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`. Omitted or blank IDs return `REQUEST_ID_REQUIRED`; invalid format returns `INVALID_REQUEST_ID`.
 
 ## Recommended Chaining
 - `+list_bi_panels` -> `+get_bi_panel_detail` -> `+query_bi_panel_data`

package/skills/ae-analysis/references/query_dashboard_report_data.md

@@ -41,20 +41,20 @@ ae-cli analysis +query_dashboard_report_data --dry-run
 | `--time_granularity` | No | Optional time granularity used to override the report default. Supported values: minute, minute5, minute10, hour, day, week, month, quarter, year, total. |
 | `--use_cache` | No | Whether to use cache. Default: true |
 | `--report_ids` | No | Optional list of report IDs. If omitted, all reports in the dashboard are queried. |
-| `--request_id` | No | Optional unique request ID used for tracking and cancellation. If provided, it must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`. For long-running or cancelable queries, provide this before starting the query so it can be cancelled later with `+cancel_query --request_id <same value>`, even if the caller stops waiting before the tool returns. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running. The auto-generated requestId is not available when the HTTP request fails before a response, so preset `requestId` is required for proactive cleanup. Generated automatically if omitted. The response `metadata.requestId` can also be passed to `cancel_query` when the query is no longer needed. |
+| `--request_id` | Yes | Required unique request ID used for tracking and cancellation. Generate it before starting the query. It must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`. Provide this before starting the query so it can be cancelled later with `+cancel_query --request_id <same value>`, even if the caller stops waiting before the tool returns. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running. `requestId` is not auto-generated for MCP query tools because the caller must know it before the response for proactive cleanup. If omitted or blank, the backend returns `REQUEST_ID_REQUIRED`; invalid format returns `INVALID_REQUEST_ID`. The response `metadata.requestId` echoes the supplied requestId and can also be passed to `cancel_query(requestId)` when the query is no longer needed. |
 | `--timeout_minutes` | No | Query timeout in minutes. If omitted, 30 minutes is used. |
 
 ## Decision Rules
 - `filters` cannot be written from experience alone: it must satisfy both the filter schema and the project metadata constraints.
 - Before calling `list_events` / `list_properties`, you must first study the corresponding reference documents.
-- On the first run, start with only the required parameters (`--project_id`, `--dashboard_id`) and add optional parameters after confirming the path works.
+- On the first run, include all required parameters (`--project_id`, `--dashboard_id`, `--request_id`) and add optional parameters after confirming the path works.
 - Wrap JSON parameters in single quotes (for example `--filters '{}'`, `--report_ids '[]'`) to avoid shell escaping issues.
 - When dates/time ranges are involved, first verify with a short range, then gradually expand the range.
-- For long-running or cancelable queries, supply your own `--request_id` before starting so `+cancel_query` can cancel by the same ID if the caller or user stops waiting. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running; call `+cancel_query --request_id <same value>` with the preset ID. The auto-generated requestId is not available when the HTTP request fails before a response. The value must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`.
+- Generate and pass `--request_id` before starting any cancelable query; it is required. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running; call `+cancel_query --request_id <same value>` with the preset ID. The value must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`. Omitted or blank IDs return `REQUEST_ID_REQUIRED`; invalid format returns `INVALID_REQUEST_ID`.
 - For cross-project troubleshooting, first confirm whether `--project_id` matches the current permissions and target environment.
 
 ## Next Steps on Failure
-- If required parameters are missing, fall back to the smallest runnable command and fill them in first (focus on `--project_id`, `--dashboard_id`).
+- If required parameters are missing, fall back to the smallest runnable command and fill them in first (focus on `--project_id`, `--dashboard_id`, `--request_id`).
 - If `Invalid JSON` appears, first check the filter schema required fields, then verify whether the event/property names come from metadata query results for the same `project_id`.
 - If the query times out or results are abnormal, first narrow the time range / grouping dimensions, then split the subqueries to locate the issue.
 

package/skills/ae-analysis/references/query_entity_details.md

@@ -42,19 +42,19 @@ ae-cli analysis +query_entity_details --dry-run
 | `--limit` | No | Optional result limit. Default: 1000, maximum: 10000                                                                                                                                                                                         |
 | `--zone_offset` | No | Time zone offset. For example, UTC+8 is 8                                                                                                                                                                                               |
 | `--use_cache` | No | Whether to use cache. Default: true                                                                                                                                                                                                     |
-| `--request_id` | No | Optional unique request ID used for tracking and cancellation. If provided, it must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`. For long-running or cancelable queries, provide this before starting the query so it can be cancelled later with `+cancel_query --request_id <same value>`, even if the caller stops waiting before the tool returns. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running. The auto-generated requestId is not available when the HTTP request fails before a response, so preset `requestId` is required for proactive cleanup. Generated automatically if omitted. The response `metadata.requestId` can also be passed to `cancel_query` when the query is no longer needed. |
+| `--request_id` | Yes | Required unique request ID used for tracking and cancellation. Generate it before starting the query. It must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`. Provide this before starting the query so it can be cancelled later with `+cancel_query --request_id <same value>`, even if the caller stops waiting before the tool returns. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running. `requestId` is not auto-generated for MCP query tools because the caller must know it before the response for proactive cleanup. If omitted or blank, the backend returns `REQUEST_ID_REQUIRED`; invalid format returns `INVALID_REQUEST_ID`. The response `metadata.requestId` echoes the supplied requestId and can also be passed to `cancel_query(requestId)` when the query is no longer needed. |
 | `--timeout_minutes` | No | Query timeout in minutes. If omitted, 30 minutes is used. |
 
 ## Decision Rules
 - `definition` / `properties` cannot be written by hand based on experience alone: they must satisfy both the schema structure and the project's real metadata.
 - `list_events` / `list_properties` must be learned from the corresponding reference docs before calling them.
-- For the first run, it is recommended to pass only the required parameters (`--project_id`, `--definition`) and add optional parameters after confirming the chain works.
+- For the first run, include all required parameters (`--project_id`, `--definition`, `--request_id`) and add optional parameters after confirming the chain works.
 - Wrap JSON parameters in single quotes (for example `--definition '{}'`, `--properties '{}'`) to avoid shell escaping issues.
-- For long-running or cancelable queries, supply your own `--request_id` before starting so `+cancel_query` can cancel by the same ID if the caller or user stops waiting. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running; call `+cancel_query --request_id <same value>` with the preset ID. The auto-generated requestId is not available when the HTTP request fails before a response. The value must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`.
+- Generate and pass `--request_id` before starting any cancelable query; it is required. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running; call `+cancel_query --request_id <same value>` with the preset ID. The value must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`. Omitted or blank IDs return `REQUEST_ID_REQUIRED`; invalid format returns `INVALID_REQUEST_ID`.
 - For cross-project troubleshooting, first confirm whether `--project_id` matches the current permissions and target environment.
 
 ## Next Step After Failure
-- If required parameters are missing, fall back to the smallest runnable command and fill them in (focus on `--project_id`, `--definition`).
+- If required parameters are missing, fall back to the smallest runnable command and fill them in (focus on `--project_id`, `--definition`, `--request_id`).
 - If `Invalid JSON` appears, first check the required schema fields, then verify that the event name/property name comes from metadata queried in the same `project_id`.
 - If the query times out or the result is abnormal, first reduce the time range/grouping dimensions, then split the query to locate the issue.
 

package/skills/ae-analysis/references/query_event_details.md

@@ -44,16 +44,16 @@ ae-cli analysis +query_event_details --dry-run
 | `--limit` | No | Optional result limit. Default: 1000 |
 | `--zone_offset` | No | Time zone offset. For example, UTC+8 is 8 |
 | `--use_cache` | No | Whether to use cache. Default: true |
-| `--request_id` | No | Optional unique request ID used for tracking and cancellation. If provided, it must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`. For long-running or cancelable queries, provide this before starting the query so it can be cancelled later with `+cancel_query --request_id <same value>`, even if the caller stops waiting before the tool returns. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running. The auto-generated requestId is not available when the HTTP request fails before a response, so preset `requestId` is required for proactive cleanup. Generated automatically if omitted. The response `metadata.requestId` can also be passed to `cancel_query` when the query is no longer needed. |
+| `--request_id` | Yes | Required unique request ID used for tracking and cancellation. Generate it before starting the query. It must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`. Provide this before starting the query so it can be cancelled later with `+cancel_query --request_id <same value>`, even if the caller stops waiting before the tool returns. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running. `requestId` is not auto-generated for MCP query tools because the caller must know it before the response for proactive cleanup. If omitted or blank, the backend returns `REQUEST_ID_REQUIRED`; invalid format returns `INVALID_REQUEST_ID`. The response `metadata.requestId` echoes the supplied requestId and can also be passed to `cancel_query(requestId)` when the query is no longer needed. |
 | `--timeout_minutes` | No | Query timeout in minutes. If omitted, 30 minutes is used. |
 
 ## Decision Rules
 - `filters` / `properties` cannot be written from experience alone: they must satisfy both the schema structure and the project metadata constraints.
 - Before calling `list_events` / `list_properties`, you must first study the corresponding reference documents.
-- On the first run, start with only the required parameters (`--project_id`, `--event_name`) and add optional parameters after confirming the path works.
+- On the first run, include all required parameters (`--project_id`, `--event_name`, `--request_id`) and add optional parameters after confirming the path works.
 - Wrap JSON parameters in single quotes (for example `--filters '{}'`, `--properties '{}'`) to avoid shell escaping issues.
 - When dates/time ranges are involved, first verify with a short range, then gradually expand the range.
-- For long-running or cancelable queries, supply your own `--request_id` before starting so `+cancel_query` can cancel by the same ID if the caller or user stops waiting. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running; call `+cancel_query --request_id <same value>` with the preset ID. The auto-generated requestId is not available when the HTTP request fails before a response. The value must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`.
+- Generate and pass `--request_id` before starting any cancelable query; it is required. If `fetch failed`, HTTP timeout, or caller timeout happens, the backend query may still be running; call `+cancel_query --request_id <same value>` with the preset ID. The value must use `mcp_<32 lowercase hex UUID>`, for example `mcp_0123456789abcdef0123456789abcdef`. Omitted or blank IDs return `REQUEST_ID_REQUIRED`; invalid format returns `INVALID_REQUEST_ID`.
 - For cross-project troubleshooting, first confirm whether `--project_id` matches the current permissions and target environment.
 
 ## Next Steps on Failure