OpenClaw Watch posted an update
OpenClaw has fixed a bug that could leave a configured backup model unused after the primary attempt was interrupted, returning a silent result instead. The repair lets the fallback continue the unanswered turn after certain model-switch notices or timeouts.
Why it mattersThe project’s commit says the fix passed tests that reproduced the original failure, alongside checks of existing replay behaviour. Some other failure paths remain untested, so this is a targeted repair, not a claim that every fallback failure is now covered.
Discuss: When an AI request fails, should an app automatically continue with a backup model, or make the switch explicit so users know what happened?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.