Flashduty brings alert management, response, and AI SRE together to help teams investigate and resolve incidents faster—and protect the trust users place in them.
Flashduty brings alert management, response, and AI SRE together to help teams investigate and resolve incidents faster—and protect the trust users place in them.
From alert to resolution — respond faster, reduce downtime, and keep customers informed.
Noise reductionGrouping, suppression, and flap detection turn dozens of duplicate alerts into one incident.
Flexible schedulesRotations, swaps, and overrides follow your team, and everyone can see who's on call.
Proactive triageChecks alerts, logs, changes, and dependencies, then suggests a likely root cause and next steps.
Summon it anywhereMention @AI SRE in any group chat or DM to keep digging. Everyone sees the same investigation.
Handle incidents in chatAcknowledge, escalate, and resolve without leaving chat. Status and the timeline sync back to Flashduty.
AI drafts the postmortemAlerts, the timeline, and the war-room chat become a structured draft in minutes.
Our status pages are now FREE for every team.
All systems are operating as expected.
API service anomaly
We've detected an issue with the API service and are investigating.
We've detected an issue with the API service and are investigating.
Customers hear from you firstAI SRE drafts customer updates as the incident unfolds. Approve one, and the status page updates with it.
Updates reach every subscriberEach update goes to the page and out to every subscribed channel.
Errors grouped into issuesSimilar errors merge into one issue, mapped back to source lines and routed to on-call.
One SDK, every platformWeb, mobile apps, and mini programs share one data model, so experience data lands in one place.
Edge engineRuns in your data center, VPC, or Kubernetes, so raw data never leaves your network.
AI writes the queryDescribe what you want to measure in plain English, and AI writes the query.
Connect 200+ alert sources in minutes, including Prometheus, Grafana, Zabbix, CloudWatch, and Datadog.
Prometheus
Grafana
Datadog
Elastic
Aliyun SLS
CloudWatch
Tencent CLS
Azure Monitor
Dynatrace
Nightingale
Loki
ClickHouse
Slack
Argo CD
CloudflarePrometheusGrafanaDatadogElasticAliyun SLSCloudWatchTencent CLSAzure MonitorDynatraceNightingaleLokiClickHouseSlackArgo CDCloudflareFrom alert noise reduction and on-call escalation to AI investigation and post-incident reviews, every step is connected—so your team can focus on restoring service.
90%+
Grouping, inhibition, flap detection, and storm warnings work in layers to condense hundreds of alerts into one incident worth acknowledging.
15 / 100
Only people who handle incidents need a seat. Notifications and collaboration are free, and most teams save 80–90% compared with PagerDuty.
L1L2L3
Flexible rotations, overrides, and coverage rules. Slack, Teams, Feishu, DingTalk, WeCom, phone, SMS, email, and the mobile app share one escalation path, with repeat reminders until someone acknowledges.
24 / 7
AI clusters alerts, recommends root causes, summarizes incidents, and drafts postmortems, so your team spends its time on recovery instead of paperwork.
200+1
Inbound alerts are deduplicated, routed, and transformed automatically, with labels enriched from your CMDB or service catalog. Built-in dashboards track MTTA, MTTR, and alert trends.
Same monitoring stack, but our P1 pages dropped from 140 a week to 11. The noise is gone, and the team finally sleeps.
We migrated off PagerDuty for a fraction of the cost — no capability lost, and on-call rotations feel humane again.
Pay only for active users and scale as needed.
AI SRE and intelligent response
Get ProfessionalEverything in Standard, plus
A custom dashboard belongs to the application it was created in, and every panel reads that application only. A Sampled mark on a chart runs that one chart over all data. The same message raised from two places makes two Issues. Android, iOS, HarmonyOS, Flutter, and React Native apps answer natural-language questions about crashes, ANR, and jank, and the Web SDK can attach response headers and request bodies to resource events.
CloudMonitor 2.0 notification policies send issue trigger, update, and recovery notifications straight to Flashduty, where one notification becomes one alert event keyed by incidentId. The BlueKing ITSM integration creates BlueKing tickets from incidents and synchronizes a linked ticket's workflow node or completion state back to the incident.
A new Allow AI to reply proactively switch, off by default, lets AI read group messages that don't mention it and reply when a reply is worthwhile, and each chat can turn it off with /ambient. When a group's team or runtime environment is unusable, AI SRE explains instead of investigating, then runs the original question again once it is fixed. One label or custom field on the incident list now takes several values.
The CLI gained flashduty safari send and watch, so a session starts from your terminal and progress arrives one JSON object per line. flashduty channel integration-* manages channel-owned integrations. The sparkle in the RUM Explorer turns a sentence into query conditions and charts and runs it, the composer predicts your next message, and a batch incident reference opens a list.
Start free, no credit card required. Or book a demo to see the full incident workflow end to end.
