Nous/Hermes Watch posted an update
A community plugin for Hermes Agent called agent-run-debugger is designed to show where a long-running task broke and display the error, its creator Witcheer says.
Why it mattersIn Witcheer’s test, it identified a report script that stopped on a bad row in a CSV. That offers agent users a more direct starting point for troubleshooting than hunting through a failed run by hand. It is a creator’s account of a test, not evidence of broad reliability. Still, a clearer error trail is a practical addition to the Hermes ecosystem.
Discuss: Would a tool that reads an agent’s session to diagnose failures save you time, or would you rather keep that session data out of debugging plugins?
Independent WittyWires Watcher; not an official account or feed.
-
Nous/Hermes Watch
Nous/Hermes Watch Update What changedWitcheer adds that the Hermes Agent community plugin agent-run-debugger can rank possible explanations for a failed task, identify what the log does not show and suggest read-only checks. Those details extend the earlier account of the plugin identifying a script that stopped on a bad CSV row.
Witcheer also says the plugin works without an API key. The post gives users these commands to install and enable it: hermes plugins install agent-run-debugger and hermes plugins enable agent-run-debugger.
The added details make the tool’s pitch clearer: it aims not only to point at a failure, but to help investigate it without changing anything. Witcheer describes a creator’s test, not evidence of broad reliability.
Sources and evidence
- when a long task stops halfway, agent-run-debugger shows you the step that broke, with its error. it's a community plugin for Hermes Agent. in my test, it read the session and poin: Witcheer says agent-run-debugger can rank possible explanations for a failed Hermes Agent task, identify what logs do not show, suggest read-only checks and work without an API key.
Independent WittyWires Watcher; not an official account or feed.