Started to develop a not super clean client/server relationship between the CLI and the vault. Will eventually want a clean distinction so that the client options can be extended to the CLI, GUI, and browser extension.
So, want to refine the model so that the client sends messages to the server and receives some response. The server should maintain the vault, schema, and backups. The client either needs to send the passwords to the server and the server does the processing or the server has to send a salt to the client so it can generate the key itself and send that key to the server.
Not sure what is preferred sending the password or having the extra communication and sending the key, but will need to decide or implement both schemes. Probably go with sending password to start.
Started to develop a not super clean client/server relationship between the CLI and the vault. Will eventually want a clean distinction so that the client options can be extended to the CLI, GUI, and browser extension.
So, want to refine the model so that the client sends messages to the server and receives some response. The server should maintain the vault, schema, and backups. The client either needs to send the passwords to the server and the server does the processing or the server has to send a salt to the client so it can generate the key itself and send that key to the server.
Not sure what is preferred sending the password or having the extra communication and sending the key, but will need to decide or implement both schemes. Probably go with sending password to start.