Run incidents where your team already is.
Opsphie is incident management that lives in Slack. One command opens a channel with everything you need.
By trying the demo, you agree to the Terms and Privacy policy.
OpsphieAPP
INC-42: Checkout returning 500s
- Severity
- P1 — High
- Status
- Identified
- Commander
- @ana
Summary
Payments fail after this morning's deploy. Rolling back.
Edit details in the web app
Declare
From “is something broken?” to a war room in seconds
Declare from where the alarm went off.
/inc newfrom any channel, with a title or without- A dedicated channel, with the commander invited
- A control panel pinned at the top for severity, status, commander and summary
lee09:11
ana09:12
Respond
A timeline that writes itself
Every change is recorded as it happens, so nobody has to take notes mid-incident and the story is ready when you need it.
- Severity, status and commander changes, updates and summaries, captured automatically
- React with 📌 or use Add to timeline to keep any message
- Add notes from the web app afterwards, placed at the time things happened
INC-42 · Timeline
- 09:12@ana declared the incident P2
- 09:15@ana changed severity from P2 to P1
- 09:21@lee: Errors started right after the 09:05 deploy
- 09:24@ana changed status from Investigating to Identified
- 09:31@ana posted an update: Rolling back the deploy.
- 09:48@ana changed status from Mitigated to Resolved
Postmortems
Postmortems built from the incident itself
Write up what happened in the web app. The incident's own details show in the postmortem as they are, so nobody copies them over.
- Templates mix written sections, each with a prompt and starter text, with ones that show the incident's summary, root cause, remediations and timeline
- A checklist of what's still empty, and a postmortem runner to get it written
- Draft, then review; publishing posts it in the incident channel and announces it
INC-42 · Postmortem
In reviewSummary
From the incidentCheckout returned 500s for 36 minutes after the 09:05 deploy.
Root cause and trigger
From the incidentA config change in the deploy pointed checkout at a retired payments endpoint.
What went well
Support flagged it within minutes, and the rollback was quick.
Checklist
- Summary
- Root cause
- What went well
- Where we got lucky
- At least one remediation
Communicate
Everyone stays in the loop
Stakeholders see what matters without joining the incident channel or asking for updates.
- New incidents, severity changes, resolutions and cancellations posted to an announcement channel you choose
- Anyone who joins the incident channel is tracked as a responder
- A web dashboard with incident history and full timelines
OpsphieAPP09:12
OpsphieAPP09:15
OpsphieAPP09:48
Learn
See whether you're getting better
Reports in the web app show how many incidents you had, how severe they were and how long they took, for any date range.
- Incidents by severity, per day, week or month
- Time to resolve (median and p90) for each severity, and time to detect and mitigate wherever those times were recorded
- Time spent investigating, identified and mitigated
- The incidents behind the numbers, one click away; test and canceled incidents are left out
- Filter the incident list by severity, status, commander and date, and share the link
Reports · Last 90 days
Time to resolve (median)
1h 12m
p90 4h 30m
Time to detect (median)
9m
From 18 incidents of 26
MCP integration
Bring your own AI!
Opsphie's MCP server lets Claude Code, Cursor or VS Code read and update incidents as you, without leaving your editor or terminal.
- Ask what's open, or for an incident's full timeline
- Post updates and change status, severity or summary, shown in Slack like any other change
- It sees only what you can see, and its changes are marked “via AI assistant”
- A personal API token you can revoke any time
> What's going on with INC-42? If the rollback fixed it, post an update.
● opsphie · get_incident(42)
INC-42 Checkout returning 500s · P1 · Mitigated
● opsphie · post_update(42)
Posted in #inc-42-checkout-errors
Errors stopped after the 09:31 rollback, so I posted an update: “Rollback is out and checkout is healthy again; monitoring for 30 minutes.”
Bring Opsphie to your workspace.
Add it to Slack, and everything starts with /inc. In an incident channel, a bare /inc opens the actions for where the incident is at.
/inc new
Declare an incident
/inc update
Post an update
/inc advance
Move it along
/inc roles
Take command
/inc edit
Keep the summary current
By trying the demo, you agree to the Terms and Privacy policy.