SWE-agent takes a GitHub issue and tries to automatically fix it, using your LM of choice. It can also be employed for offensive cybersecurity or competitive coding challenges. [NeurIPS 2024]
Fix: Do not use default settings with run_replay
Kilian Lieret0a41bc476570
diff --git a/sweagent/run/run_replay.py b/sweagent/run/run_replay.pyindex 85cc9bc1..af9a1dc9 100644--- a/sweagent/run/run_replay.py+++ b/sweagent/run/run_replay.py@@ -207,7 +207,7 @@ def run_from_cli(args: list[str] | None = None):help_text = ( # type: ignore__doc__ + "\n[cyan][bold]=== ALL THE OPTIONS ===[/bold][/cyan]\n\n" + ConfigHelper().get_help(RunReplayConfig))- run_from_config(BasicCLI(RunReplayConfig, help_text=help_text).get_config(args)) # type: ignore+ run_from_config(BasicCLI(RunReplayConfig, help_text=help_text, default_settings=False).get_config(args)) # type: ignoreif __name__ == "__main__":