Skip to content

Firefight can write the first draft of a postmortem for you. The AI reads the incident record and produces a structured document in the nine standard sections, written in a blameless tone that focuses on systems and processes rather than individuals.

When an incident is resolved, the resolution message in its channel offers a Write the postmortem button. You can also run /ff postmortem in the channel. Either works only once the incident is resolved and while it has no postmortem yet. Firefight replies that the draft is being written, which usually takes under a minute.

On the incident page in the web dashboard, the postmortem card offers drafting once the incident is resolved. Click Generate draft. You land on the postmortem page, which shows a placeholder while the draft is written and switches to the editor as soon as it is ready.

Firefight posts a Postmortem message in the incident channel and pins it, so the channel always has it to hand. The message carries the postmortem’s title, how long the incident lasted, its severity and its lead, with a View full postmortem button that opens it in the dashboard. A blank postmortem you start yourself is not posted.

The AI works only from what Firefight recorded about the incident. That includes:

InputExamples
Incident detailsName, severity, status, who declared it, who led it, key timestamps, duration
Custom fieldsAny values your team filled in on the incident forms
TimelineThe recorded events from declaration through close, with the fields each update changed
Status updatesThe full text of every update posted with /ff update, from the dashboard, or by an agent
Channel discussionA narrative summary of what people wrote in the incident channel
Actions and follow-upsWhat was tried during response and what remains open, with links to any tracked issue
ShoutoutsRecognition given during the incident

This is why disciplined use of /ff update, actions, and the incident channel pays off. The richer the record, the better the draft. An incident run entirely through posted updates, with nobody typing in the channel, still gives the AI everything those updates said.

Firefight writes two sections itself rather than leaving them to the AI. The Timeline lists every recorded event. Under each update it shows what changed, such as Severity: Minor → Major, and the update’s first line. The full text stays on the incident’s own timeline.

Action items starts with every follow-up already on the incident, each with its status, who it is assigned to while it is still open, and a link to its issue when it has one. Any further items the AI draws from the causes in the record follow underneath.

The AI writes only what the timeline, the status updates and the channel messages support. It never guesses a cause, an impact or a fix. Every heading still appears. The Timeline is always filled in, and under any heading it had nothing for you find a short note asking you to add what you know. Post in the incident channel while it runs and the draft has more to work from.

If the draft cannot be written, the postmortem page says so and offers Try again and Start blank. The person who asked for the draft is also told in the incident channel. When Firefight restarted while writing the draft and could not finish it, the page says that instead, with the same two choices. When Halon cannot write it because the workspace has no AI account set up or its credit has run out, the page says which, and who can fix it.

POST /api/v1/incidents/:id/postmortem with generate: true starts the same draft, and start_postmortem does it over MCP. Drafting takes a moment, so the response comes back straight away with a generation_state. Read the postmortem back until that state clears, then edit it like any other.

Leave generate out and you get the blank one described below.

If you prefer to write from scratch, click Start blank on the incident page card. You get an empty document in Draft status with the incident’s identifier and name as the title. Starting blank does not use AI.

Inside the editor you can point AI at any passage. Select some text and click the sparkles button in the floating toolbar. A Rewrite with AI dialog opens where you describe what you want, for example “make this more concise” or “expand on the root cause”. The AI has access to the incident context, including the timeline, summary, and actions, so it can add real detail rather than padding. The rewritten text replaces your selection, and you can undo it like any other edit.

Treat the generated document as a strong first draft, not a finished artifact. The AI only knows what made it into the incident record. It cannot know about the hallway conversation, the dashboard someone eyeballed, or the context behind a decision. Before you mark a postmortem completed, have someone who was on the incident verify the facts, sharpen the contributing factors, and make sure every action item has an owner.