Historical Chat

Sometimes you don't want a dashboard — you want an answer. "Which of our PRs this quarter were actually risky?" shouldn't require remembering which page has that chart and how to filter it. Historical Chat lets you just ask, in plain language, and get an answer built from your project's analyzed history — with the receipts attached.

Asking questions

Type a question the way you'd ask a colleague. Firetrail answers from your analyzed PR history, and its answers cite their sources: clickable PR links and the relevant numbers, so you can verify anything it claims. Questions that work well:

• "What are the highest risk PRs?"
• "Which bugfixes were caused by risky PRs?"
• "Who is missing tests?"
• "Which PRs mixed dependency updates with features?"
• "Give me an overview of engineering health."

Anything Firetrail measures is fair game: scores and bands, classifier findings, test coverage, bugfix origins, authors, and time ranges.

Where it shines

The moments when a dashboard is too slow:

• In a retro — "show me this sprint's high-risk PRs" while everyone's looking.
• In planning — "what usually goes wrong when we touch the payments service?"
• When a stakeholder asks something oddly specific — and you'd rather answer in thirty seconds than build a spreadsheet.

One honest limit

The chat knows what Firetrail has analyzed — nothing else. It doesn't see un-analyzed PRs (run a sync to close gaps), and it can't answer questions about things Firetrail doesn't measure. For the standard views of the same data, Engineering Health and Team Comparison are a click away.

Availability

Historical Chat is included on Pro and Business plans. See Plans & billing.