ysyneu · Oct 20, 2026 · Oct 20, 2026
diff --git a/en/ai-sre/agents.mdx b/en/ai-sre/agents.mdx
@@ -20,7 +20,7 @@ sidebarTitle: Agent
 
 Delegation doesn't make you wait. After AI SRE delegates a task to a remote agent, the conversation continues immediately. You can keep working or delegate several tasks at once. When the remote agent finishes, its result appears in the conversation as a new message.
 
-Each A2A delegation appears in the conversation stream as one line: a robot icon, "Dispatched" plus the remote agent name, and the task intent. On the right it shows the run status and usage metrics such as tool call count, tokens, and elapsed time. The run status is one of: initializing, in progress, awaiting input, awaiting authorization, completed, failed (including orphaned), interrupted, or killed. Click the line to view the full delegation trace in the right-hand panel; the `A2A` badge sits in that panel's title, not on the line.
+Each A2A delegation appears in the conversation stream as one line: a robot icon, "Dispatched" plus the remote agent name, and the task intent. On the right it shows the run status and usage metrics such as tool call count, tokens, and elapsed time. The run status is one of: initializing, in progress, awaiting input, awaiting authorization, completed, failed (including orphaned), interrupted, or killed. Click the line to view the full delegation trace in the right-hand panel; the `A2A` badge sits in that panel's title, not on the line. A delegation running on a BYOC Runner survives a dropped connection or an AI SRE instance restart (another instance takes it over), while a delegation on a cloud environment ends with its instance — so the interrupted and failed (including orphaned) statuses mostly appear on cloud environments.
 
 <Tip>
 **The instructions decide whether AI SRE picks this agent.** Before delegating, AI SRE sees only one `name: summary` line per agent and uses it to decide. The summary comes from the `summary:` line at the top of the instructions. The body is delivered in full only on the first delegation, so it can be long and detailed. See [Writing Instructions](#writing-instructions).

diff --git a/en/ai-sre/artifacts.mdx b/en/ai-sre/artifacts.mdx
@@ -37,8 +37,8 @@ Entry point: **AI SRE → Artifacts** in the left navigation.
 | Form | What you see |
 |---|---|
 | Multi-file page | One HTML page can bundle its styles, scripts, and images and is shown as-is on the artifact detail page. A single publish allows up to 255 files (including the entry page) and 64 MiB in total. A page, text, data, or code file can be up to 16 MiB each; images, video, audio, PDF, and archives up to 15 MiB each |
-| Canvas | Several pages laid out on one canvas that you can pan and zoom (2%–400%); select a board to view it full window. Suited to reviewing several interface designs side by side |
-| Slides | Each page is a fixed 1920×1080. Turn pages with the arrow keys or Space, and select **Present** for full screen |
+| Canvas | Several pages laid out on one canvas that you can pan and zoom (2%–400%); select a board to view it full window, where ←/→ or the **Previous** / **Next** buttons step through every board in the canvas and Esc or **Close** exits. Suited to reviewing several interface designs side by side |
+| Slides | Each page is a fixed 1920×1080. Turn pages with the arrow keys, PageUp / PageDown, Home / End, or Space (Shift + Space goes back), and swipe left or right on a touch screen; select **Present** for full screen, where clicking the left or right half of the screen turns the page and the page number shows at the bottom |
 | Table | `.csv` / `.tsv` files display as a table with a fixed header row. Up to 1000 rows and 200 columns are shown, within 20000 cells in total, so a wide table shows fewer rows |
 | Video & audio | `.mp4` `.webm` `.mov` and `.mp3` `.m4a` `.wav` play in the browser's built-in player |
 
@@ -56,7 +56,7 @@ You can't upload files to the artifact gallery directly. The agent produces and
 - **Paging**: 12 / 24 / 48 per page, 24 by default.
 - **Shareable URL**: the page URL keeps the scope, search term, sorting, and page. Copy it to share or bookmark the filtered list. When you return from a detail page, you land where you were.
 
-Each card shows a content thumbnail, a type icon, the title, the edited time, and the scope. It also shows whether a public link is on and whether the artifact is pinned. The thumbnail is a static preview for HTML, Markdown, images, and plain text, and an empty frame for PDFs and other types. Hover over the edited time to see the exact time.
+Each card shows a content thumbnail, a type icon, the title, the edited time, and the scope. It also shows whether a public link is on and whether the artifact is pinned. The thumbnail is a static preview for HTML, Markdown, images, plain text, tables (the header and first few rows), video (the first frame), and canvases / slide decks (their first board or first slide); only PDFs and audio show an empty frame. Hover over the edited time to see the exact time.
 
 ### Pinning
 
@@ -133,10 +133,6 @@ Visitors see a read-only page marked "Content is user-generated and unverified."
 | Update snapshot | Appears only when the artifact's content has changed since the snapshot was taken (an "Artifact content has changed — update the snapshot to sync the latest content" notice). Clicking it overwrites the snapshot with the latest content; **the link itself stays the same** |
 | Revoke public link | Turns off public access; the link stops working immediately |
 
-<Note>
-Artifacts larger than **16 MiB** can't generate a public link — you'll see a notice that the artifact exceeds the size limit. You can still share it using "Account only" mode.
-</Note>
-
 ## Update a published artifact
 
 Publishing the same file again does **not** silently overwrite a published artifact. When the agent updates an artifact, it names the artifact and the version it is based on. If someone changed the artifact in the meantime, the agent reads the current content and merges before retrying, instead of overwriting it. The link and artifact ID stay the same after an update.

diff --git a/en/ai-sre/automations.mdx b/en/ai-sre/automations.mdx
@@ -222,7 +222,7 @@ In that analysis's chat, a merge appears as one process line: `Joined this analy
 Two platform limits are fixed and not configurable:
 
 - **One analysis takes in at most 20 incidents.** Once it is full it stops receiving similar incidents, and later ones start a new run, which carries on absorbing.
-- **A finished run can only be joined within 10 minutes of finishing.** A run still `running` or `retrying` is always joinable. A run that finished well (`succeeded` / `partial`) stays joinable for 10 minutes after it completed, and after that a similar incident starts a new run. Failed, skipped, and abandoned runs are never joined.
+- **A finished run can only be joined within 60 minutes of finishing.** A run still `running` or `retrying` is always joinable. A run that finished well (`succeeded` / `partial`) stays joinable for 60 minutes after it completed, and after that a similar incident starts a new run. Failed, skipped, and abandoned runs are never joined.
 
 <Note>
 The similarity comparison requires the account's AIOps capability and a Pro or higher subscription. If AIOps is off or the edition doesn't qualify, no comparison runs, and every incident starts its own run.

diff --git a/en/ai-sre/environments.mdx b/en/ai-sre/environments.mdx
@@ -420,7 +420,7 @@ When troubleshooting Runner issues, start with the **Status** and **Last heartbe
 
 <AccordionGroup>
 <Accordion title="Runner shows Offline" icon="bug">
-  Offline means the Runner connected before, but AI SRE has not received a heartbeat for about 90 seconds. Check whether the systemd service or process is running, whether the machine is sleeping or offline, and whether outbound traffic to AI SRE is blocked by a firewall. After process and network recovery, the Runner reconnects automatically.
+  Offline means the Runner connected before, but AI SRE has not received a heartbeat for about 90 seconds. Check whether the systemd service or process is running, whether the machine is sleeping or offline, and whether outbound traffic to AI SRE is blocked by a firewall. After process and network recovery, the Runner reconnects automatically; any A2A delegation still in progress then carries on (the Runner retains and replays what it has already produced) rather than failing or rerunning the whole round.
 </Accordion>
 <Accordion title="Runner stays Pending" icon="wrench">
   Pending means this Environment has never connected successfully. Confirm that the install command completed, the Token belongs to this Environment, and the `URL` is reachable from the target machine. If you configured permissions, also confirm that the file exists and the YAML parses.

diff --git a/en/ai-sre/sessions.mdx b/en/ai-sre/sessions.mdx
@@ -216,6 +216,8 @@ Click a file the agent produced to preview it on the right:
 | Code | Syntax highlighting and line numbers |
 | Markdown / HTML | Rendered view, with a **Source** toggle |
 | Images, PDF | Shown directly |
+| Table (`.csv` / `.tsv`) | Shown as a table with a fixed header row. Up to 1000 rows and 200 columns are shown, within 20000 cells in total, so a wide table shows fewer rows |
+| Video & audio (`.mp4` `.webm` `.mov` / `.mp3` `.m4a` `.wav`) | Play in the browser's built-in player; if a file can't be played, you're asked to download it |
 | Skill package (`.skill`) | File tree and contents; download it or **Save Skill** to your account |
 
 Previews support **Copy** and **Download**. All artifacts can also be managed on the **Artifacts** page in the left navigation — see <a href="/en/ai-sre/artifacts">Artifacts</a>.

diff --git a/en/on-call/channel/create-edit.mdx b/en/on-call/channel/create-edit.mdx
@@ -124,7 +124,7 @@ Configure rules in **Configuration** → **Integrate Data** → **Drop Rules** t
 
 <Steps>
   <Step title="Create a Drop Rule">
-    Click **Add Drop Rule**.
+    Click **New rule** in the upper-right corner of the page. The drop rule list is a table: click any row to expand it and view that rule's drop conditions, and columns such as **Last modified** can be shown or hidden from the **View** menu.
   </Step>
   <Step title="Fill in Rule Information">
     Enter a **Rule Name** (required) and **Description** (optional).

diff --git a/en/on-call/channel/escalation-rule.mdx b/en/on-call/channel/escalation-rule.mdx
@@ -9,7 +9,7 @@ Escalation rules determine **when**, **through which channels**, **to whom**, an
 
 An escalation rule contains six core elements. The system matches rules from top to bottom, **stopping after the first successful match**.
 
-You can enable or disable individual escalation rules. Flashduty skips disabled rules during matching, so they trigger no notifications. You can also copy an escalation rule to the current channel or another channel to quickly reuse existing configurations; see [Copy an Escalation Rule](#copy-an-escalation-rule).
+You can enable or disable individual escalation rules. Flashduty skips disabled rules during matching, so they trigger no notifications. You can also copy an escalation rule to the current channel or another channel to quickly reuse existing configurations; see [Copy an Escalation Rule](#copy-an-escalation-rule). **Edit**, **Copy**, **Enable**/**Disable**, and **Delete** all live in the **More actions** (⋯) menu at the end of each row: an enabled rule offers **Disable**, and only after it is disabled does the menu offer **Enable** and **Delete**. **Disable** and **Delete** both ask you to confirm first.
 
 ### 1. Trigger conditions
 
@@ -22,6 +22,8 @@ Determines which incidents trigger the current rule.
   - **Specific Time Period**: for example, only active on weekdays (Monday-Friday 9:00-18:00)
   - **Service Calendar**: Reference a predefined [Service Calendar](/en/on-call/configuration/service-calendar), for example, only active on "Trading Days", automatically excluding holidays
 
+The status dot before each rule name in the list shows whether that rule is participating in matching right now: green **Enabled**, gray **Disabled**, and yellow **Outside active hours**. Yellow means the rule itself is enabled but currently falls outside its effective time, so it is skipped for now; it turns green again automatically once the effective time begins, and you don't need to enable it manually.
+
 **Incident Filter**: Controls which incidents the rule matches.
 
 - **Off (Default)**: All incidents may match this rule
@@ -136,7 +138,7 @@ The minimum escalation timeout is 1 minute.
 
 ## Copy an escalation rule
 
-Click the **Copy** icon on an escalation rule card to set the new rule's name, target channel and description:
+In the escalation rule list, click the **More actions** (⋯) icon at the end of a row and select **Copy** to set the new rule's name, target channel and description:
 
 | Field | Description |
 | :--- | :--- |

diff --git a/en/on-call/channel/noise-reduction.mdx b/en/on-call/channel/noise-reduction.mdx
@@ -348,7 +348,7 @@ When you create a quick silence, the drawer lists the labels Flashduty excluded
 
 ### Copy a silence rule
 
-Click the **Copy** icon on a silence rule card (not available for rules created by **Quick Silence**) to set the new rule's name, description and target channel:
+Silence rules are listed in a table; click a row to expand it and see the rule's details. In that list, open **More actions** (⋯) on the target rule and choose **Copy** (not offered for rules created by **Quick Silence**) to set the new rule's name, description and target channel:
 
 | Configuration | Description |
 | :--- | :--- |

diff --git a/en/on-call/integration/alert-integration/routing-rules.mdx b/en/on-call/integration/alert-integration/routing-rules.mdx
@@ -36,7 +36,7 @@ For the complete syntax and examples, see [Configure filter conditions](/en/on-c
 | Mode | Description | Use Case |
 | :--- | :--- | :--- |
 | **Standard Routing** | Manually select target channel | Alert-to-channel mapping is fixed |
-| **Name Mapping** | Map specified label value to channel name, auto-deliver | Alerts already carry business identifier and channel name matches label value |
+| **Name Mapping** | Map specified label value to channel name, auto-deliver; separate multiple names in the label value with commas | Alerts already carry business identifier and channel name matches label value |
 
 <Warning>
 When using name mapping mode, if corresponding channel doesn't exist, delivers to default route or discards.
@@ -111,6 +111,8 @@ You can perform the following actions:
 
 **Result**: Alerts with `Labels.team=payment` auto-deliver to channel named `payment`.
 
+**Multiple channels**: You can separate multiple channel names in the label value with commas; empty and duplicate items are ignored. For example, alerts with `Labels.team=payment,order` deliver to both the `payment` and `order` channels. At most 100 names are resolved in a single lookup, and names beyond that are treated as unmatched.
+
 <Frame>
 <img src="https://docs-cdn.flashcat.cloud/images/png/6e42dbbbf0ebfc9d3adb9061718b28e0.png" alt="Name mapping mode" />
 </Frame>
@@ -137,7 +139,7 @@ Confirm reported alerts can hit configured routing rules. We recommend configuri
 </Accordion>
 
 <Accordion title="When flow control is set to 'Stop', will default route still match?">
-No. The default route is independent of flow control: it only takes effect when no rule actually delivers the alert to any channel. As long as a matched rule delivers the alert to at least one channel (whether flow control is set to 'Continue' or 'Stop'), the default route will not be triggered. Note that if a name mapping rule matches but the mapped channel does not exist (so nothing is actually delivered), the default route will still be triggered.
+No. The default route is independent of flow control: it only takes effect when no rule actually delivers the alert to any channel. As long as a matched rule delivers the alert to at least one channel (whether flow control is set to 'Continue' or 'Stop'), the default route will not be triggered. There are two exceptions: if a name mapping rule matches but the mapped channel does not exist (so nothing is actually delivered), the default route will still be triggered; and if the label value of a name mapping rule lists multiple channel names and some of them have no matching channel (or more than 100 names are truncated), the default route is delivered alongside the channels that rule already matched.
 </Accordion>
 </AccordionGroup>
 

diff --git a/en/on-call/statuspage/subscriptions.mdx b/en/on-call/statuspage/subscriptions.mdx
@@ -38,7 +38,7 @@ When a visitor enters an email address on a public status page, the system sends
 </Steps>
 
 <Note>
-If the visitor changes their subscription scope before confirming, the system updates the pending subscription content without sending a new confirmation email.
+When a pending subscription already exists for the same email address, submitting the form again neither changes the scope awaiting confirmation nor sends another confirmation email — the scope chosen on the first submission is what takes effect. To change it, resubmit after the confirmation email expires (5 minutes). A confirmed subscriber who changes scope must resubmit the form and click the link in the new confirmation email: their existing subscription stays as it is until they confirm, and the new scope replaces it once confirmed. Alternatively, they can use the manage token in any notification email to update it directly.
 </Note>
 
 ### IM subscription (internal status pages)

diff --git a/en/platform/permission-design.mdx b/en/platform/permission-design.mdx
@@ -182,8 +182,9 @@ In addition to preset roles, you can create custom roles for more granular permi
 <Note>
 - System preset roles (Admin, Responder, Viewer) cannot be modified or deleted
 - Custom roles support editing, copying, enabling/disabling, and deletion
+- Disabling a custom role revokes what it granted immediately: both API authorization and the visibility of buttons and menus in the console stop granting it. The role's member bindings are kept, so re-enabling it restores its permissions
 - You can't delete a custom role that any saved SSO configuration (SAML2.0 / OIDC / CAS) uses as its **default role**. Force delete doesn't help, because it only removes member bindings. First deselect it as the default role in that SSO configuration. This applies even when SSO or role sync is off
-- A member can hold multiple roles simultaneously; their effective permissions are the union of all assigned roles
+- A member can hold multiple roles simultaneously; their effective permissions are the union of the permissions of all their **enabled** roles
 - When you [invite a member through the Open API](/en/api-reference/platform/members/member-invite) or SSO creates a member automatically without a role, the member gets **Viewer** by default
 </Note>
 

diff --git a/zh/ai-sre/agents.mdx b/zh/ai-sre/agents.mdx
@@ -20,7 +20,7 @@ sidebarTitle: Agent
 
 委派后无需等待：AI SRE 把任务交给远端 Agent 后会立即继续当前对话，您可以继续工作或同时委派多个任务；远端完成后，其结果作为一条新消息出现在对话中。
 
-每次 A2A 委派在对话流里以一行呈现：机器人图标 + “已派发〈远端 Agent 名〉”文案 + 本次任务意图，右侧显示运行状态（初始化 / 进行中 / 等待输入 / 等待授权 / 完成 / 失败（含孤儿任务）/ 中断 / 已终止）以及工具调用数、Token、耗时等用量；点击这一行可在右侧面板里查看该次委派的完整过程，`A2A` 徽标标在该面板标题上，不在对话流这一行上。
+每次 A2A 委派在对话流里以一行呈现：机器人图标 + “已派发〈远端 Agent 名〉”文案 + 本次任务意图，右侧显示运行状态（初始化 / 进行中 / 等待输入 / 等待授权 / 完成 / 失败（含孤儿任务）/ 中断 / 已终止）以及工具调用数、Token、耗时等用量；点击这一行可在右侧面板里查看该次委派的完整过程，`A2A` 徽标标在该面板标题上，不在对话流这一行上。在 BYOC Runner 上执行的委派遇到连接中断或 AI SRE 实例重启也会接着跑（由另一实例接管），只有云端环境上的委派随实例一起中断，因此「中断」与「失败（含孤儿任务）」主要出现在云端环境。
 
 <Tip>
 **调用说明决定 AI SRE 会不会选中这个 Agent。** 委派前，AI SRE 只看到清单里的一行 `名称：简介`（简介取自调用说明开头的 `summary:`），据此判断要不要派给它；正文只在首次委派时完整送达，可以写长写细。写法见[调用说明的写法](#调用说明的写法)。