Hi @google,
Developers often return to the same agent project with requirements and decisions already established. Would a small, opt-in Memcode example help show how agents-cli users can explicitly save and retrieve those approved notes across separate runs?
The example would keep memory separate from local CLI state, never upload shell history, command output, or credentials, and explain how to disable it. Hosted authentication would use the normal OAuth flow rather than custom token handling.
If this is a fit, I can prepare a concise docs or example PR.
Hi @google,
Developers often return to the same agent project with requirements and decisions already established. Would a small, opt-in Memcode example help show how agents-cli users can explicitly save and retrieve those approved notes across separate runs?
The example would keep memory separate from local CLI state, never upload shell history, command output, or credentials, and explain how to disable it. Hosted authentication would use the normal OAuth flow rather than custom token handling.
If this is a fit, I can prepare a concise docs or example PR.