AI SRE is available to accounts on an On-call Pro or higher subscription, with no application needed. Starting September 16, 2026, 08:00 Beijing time, it must be activated in the console before use and is billed on actual usage; see billing terms.
What Is AI SRE
AI SRE is Flashduty’s autonomous SRE agent platform. You give the AI instructions through conversation, and it independently investigates incidents, diagnoses root causes, calls tools to run diagnostics, and captures the operational knowledge gained from each investigation for future reuse. It is not a simple Q&A chatbot — it is a hands-on troubleshooting worker: it plans its own steps, reads and writes files, queries monitoring and logs, executes commands, calls external tools (MCP), and delegates subtasks when needed, ultimately delivering conclusions backed by a full investigation trail. AI SRE is deeply integrated with Flashduty’s incident response system. When Flashduty generates an incident or opens a war room, you can directly trigger an AI SRE session so the agent enters the investigation with full incident context — you can collaborate either in the console or by @ mentioning it directly in your IM group (Slack / Feishu / DingTalk / WeCom).
Conversation as Troubleshooting
Describe the problem in natural language. The agent plans autonomously, calls tools, and delivers an investigation trail and conclusion — no manual scripting required.
Integrated with Incident Response
Launch a session with one click from an incident or war room. The agent enters the investigation with full incident context, and the knowledge it accumulates feeds back into the next response.
Typical Use Cases
AI SRE is more than a chat box in the console — it covers multiple collaboration entry points across the full lifecycle from incident trigger to retrospective:
Conversational Troubleshooting
Ask questions proactively in the console’s Chat workspace: why a service is behaving abnormally, the root cause of an alert, or the blast radius of a change. The agent streams its planning, tool calls, and intermediate findings before delivering a final conclusion.
On-Demand in IM
No context switching needed — in Slack, Feishu, DingTalk, or WeCom group chats or direct messages, @ AI SRE to start or continue an investigation. It replies in-thread so all team members stay in the loop. See IM Integration.
Automatic War Room Diagnosis
When a war room is opened for an incident, AI SRE automatically runs an initial diagnostic and posts the conclusion back to the war room — by the time your team starts investigating, a first-pass analysis is already in the channel.
Usage Insights
Use
/insight to review the past 30 days of sessions, quantify where you’re spending time, identify missing runbooks and repeatedly copy-pasted context, and receive actionable improvement suggestions.Activation and billing
AI SRE is available to accounts holding a valid (enabled) On-call Pro or higher subscription, with no application needed.
Billing
Billing
AI SRE is billed on actual usage starting September 16, 2026, 08:00 Beijing time, settled in credits (1 credit = ¥1); usage before that date isn’t charged. Metered items are model usage, sandbox online time, and web search. Activation is free, eligible accounts get gift credits each billing period, and you can also buy credit packages (1,000 / 5,000 / 20,000 credits for ¥950 / ¥4,500 / ¥17,000, valid 365 days from payment). Pay-as-you-go is off by default; an admin can turn it on and optionally set a per-period cap. Full terms: AI SRE Credit Package Purchase Agreement.When the gift credits, credit packages, and wallet balance are all used up, AI SRE refuses new conversation requests: the composer is disabled and a non-dismissible notice stays pinned above it (the text comes from the server and covers all four shapes — “not activated yet”, “turned off”, “this period’s allowance is spent”, and “the wallet is empty”), with a Go to AI SRE billing button that opens the billing center (
/wallet/plan?product=ai-sre). An admin turns on pay-as-you-go, buys a credit package, or tops up the wallet there; once the state recovers, you can continue the conversation.Activating and closing AI SRE
Both directions are self-service on the same card in the console: open the billing center from the left navigation (Payment → Plan, the page titled “Your Plans”) and select the AI SRE tab — that is/wallet/plan?product=ai-sre, the same destination as the Go to AI SRE billing entry in the credits notice. The billing center is available on SaaS only, so On-Premises deployments have no such tab.
Opening the page requires Payment Read. The Enable AI SRE and Disable AI SRE buttons additionally require wallet purchase permission (the console’s own wording next to the activation entry is “Wallet purchase permission required”); without it the buttons are disabled and hovering shows “You are not authorized for this operation”.
Disable AI SRE first opens a confirmation dialog, “Disable AI SRE?”, whose body reads “After you disable it, this account can no longer start AI SRE investigations. Purchased credit packages and gift credits are retained, and you can enable it again at any time.” Once you confirm, the console polls the order status and reports one of the following:
What disabling changes:
- No new investigations: after you disable it, this account can no longer start AI SRE investigations, and a refused request gets the server’s own sentence, “AI SRE is turned off for this account. Turn it back on to continue.” (the credits overview reports
blocked_reason=turned_off). A top-level run that has already started is not interrupted, and the usage it goes on to produce is still metered. - Credits and packages are retained: neither the gift credits nor the balance and validity of purchased credit packages are cleared by disabling the service, and they are usable again after you re-enable it (gift credits themselves remain usable only within the billing period they were granted for, per the Agreement). Re-enabling is simply running Enable AI SRE again — nothing else is required.
- Nothing is charged or refunded: disabling itself costs nothing and triggers no refund; package refunds still follow the AI SRE Credit Package Purchase Agreement.
- Arrears and a lapsed subscription do not block it: disabling checks neither the wallet balance nor a valid On-call subscription, so an account in arrears or with a lapsed subscription can still turn the service off.
When the subscription lapses
Every request re-checks the account’s On-call subscription, so a subscription that lapses pauses AI SRE immediately. The interface shows the server’s own sentence, and distinguishes two cases whose remedies differ — don’t apply the wrong one:
The notice also appears in the console chat page (pinned above the composer, with a Go to AI SRE billing entry on its right), in IM sessions, in automation run records, and in A2A call errors; the credits overview reports the same sentence as
blocked_reason=no_license. Recovery needs no re-activation: once the renewal or upgrade lands, the next request is admitted automatically.
Core Capabilities
AI SRE is built around “conversational troubleshooting + knowledge accumulation + autonomous execution,” with every capability configurable and manageable from the console.
Conversational Troubleshooting
Collaborate with the agent session by session. A session handles your messages one at a time, with streaming output, cancellation at any time, automatic context compaction for long conversations, and one-click launch from an incident.
IM Platform
@ the agent in Slack / Feishu / DingTalk / WeCom to start or continue an investigation, and receive automatic initial diagnostics in incident war rooms.
Skills
Skill packages that the agent can invoke, encapsulating reusable troubleshooting workflows. Scope can be set to Shared (account-level) or Team; once enabled, they are loaded on demand during sessions.
Manage Knowledge
Knowledge indexed via DUTY.md as the entry point and @-reference links, carrying service catalogs, runbooks, on-call routing, and other long-lived context — loaded in layers by account and team.
MCP (External Tools)
Connect external tools and data sources via the Model Context Protocol. MCP servers are not pre-connected; the agent establishes, uses, and closes connections on demand when making a call.
Agent
Delegate tasks to external remote agents via the standard A2A protocol. AI SRE also exposes its own Agent Card so external clients can invoke it in reverse.
BYOC
The execution layer for the agent: uses Flashduty-managed cloud sandboxes by default, or deploy a persistent Runner on your own machine to let investigations reach your private network.
Usage Insights
Use /insight to review the past 30 days of AI SRE sessions and receive a quantified overview, narrative summary, and actionable operational improvement suggestions (read-only; nothing is applied automatically).
Automations
Have AI SRE run a hidden session on a cron schedule, via an API call, or on an On-call incident to automatically produce periodic inspections, insights, or post-mortems.
Apps
Authorize external applications (currently GitHub) so AI SRE can work directly inside your code repositories: understand code, investigate PRs and commits, and open a PR or file an issue on request.
Console Navigation
After entering AI SRE, the left sidebar has four entries (menu names match the console):
The three groups under Customize:
Visibility of each area is determined by your access permissions in the account: menus or sub-tabs you do not have permission for will not appear in the navigation.
Quick Start
For the full walkthrough from activating access to finishing your first investigation, see Quickstart.
Next Steps
Console
Learn about sessions, streaming output, cancellation and context compaction, and how to launch an investigation from an incident.
IM Platform
@ the agent in Slack / Feishu / DingTalk / WeCom, and learn about automatic war room diagnostics.
Usage Insights
Use /insight to review the past 30 days of sessions and surface operational friction such as repeated context and missing runbooks.