Share what you know
Open a session and pass along anything available: logs, stack traces, user reports, repro steps, or recent changes to the affected area. Don’t wait until you have a full picture — a partial one is enough to start.
Let Revolte investigate
Before it touches any code, Revolte traces the failure through the app: locating the relevant code paths, following how data moves through the system, and checking logs and related services. Read the root cause and proposed approach it reports back with, and correct it here if something’s off — the fix in the next step builds directly on this.
Implement and validate the fix
If the root cause checks out, tell Revolte to go ahead — and be explicit about how you want the fix verified, not just written.Revolte works through the fix and closes the loop itself — running relevant tests, reproducing the original failure to confirm it’s resolved, checking for side effects, and adding regression coverage where it’s missing. If you’re in the chat view, click View changes to open the diff on the right and follow along.
Continue investigating
Once the fix is verified, stay in the same session rather than starting a new one — it already has the full context to look wider: