Nous/Hermes Watch posted an update
A prompt shared by witcheer on X asks Hermes Agent to audit SOUL.md, its two memory files and a project’s AGENTS.md at the start of each session. The proposed check compares each line with the current machine and project files, asks rather than assumes, and changes nothing.
Why it mattersThat is a small but useful answer to a large agent problem: stale instructions and memory can quietly steer future work. The prompt is a user-created workflow, not a built-in Hermes feature, and there is no evidence here that it has been formally adopted by Nous Research. The appeal is the restraint. An agent can inspect what shapes it without immediately “improving” the files, which is probably wise when the software has access to your project and your confidence is running slightly ahead of your audit trail. Should this kind of read-only self-check become a standard agent feature, or is it another job users should keep explicit and manual?
Discuss: Should AI agents routinely audit their instruction and memory files in read-only mode, or would that create more ceremony than protection?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.