You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
It should be possible for users to accept or decline the proposed changes made by the AI. Additionally, users should be able to specify which files or parts of the code should remain unchanged. Currently, the AI sometimes overwrites functional code with replacements that are not desired, leading to loss of stable functionality.
Key Requirements:
Granular Control: Users should have the ability to selectively approve or reject changes at a file or even a line level.
File-Level Locking: Users should be able to mark specific files or segments as "do not modify" to protect critical or well-functioning parts of the code.
Change Review Interface: Introduce a review mechanism where users can inspect proposed changes and provide feedback or approval.
Rationale:
The current approach can lead to unwanted modifications, especially when the AI suggests updates for stable parts of the code. By giving users more granular control, we ensure that only the necessary changes are integrated, reducing the risk of breaking existing functionality.
This feature would significantly improve developer trust and control over AI-generated changes, leading to more efficient collaboration between users and the AI system.
Possible Implementation Ideas:
A dedicated UI for reviewing and approving changes similar to Cline.
A "lock" feature that prevents specific files from being altered by the AI.
Integration with version control systems to provide an additional layer of safety before applying changes.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
It should be possible for users to accept or decline the proposed changes made by the AI. Additionally, users should be able to specify which files or parts of the code should remain unchanged. Currently, the AI sometimes overwrites functional code with replacements that are not desired, leading to loss of stable functionality.
Key Requirements:
Granular Control: Users should have the ability to selectively approve or reject changes at a file or even a line level.
File-Level Locking: Users should be able to mark specific files or segments as "do not modify" to protect critical or well-functioning parts of the code.
Change Review Interface: Introduce a review mechanism where users can inspect proposed changes and provide feedback or approval.
Rationale:
The current approach can lead to unwanted modifications, especially when the AI suggests updates for stable parts of the code. By giving users more granular control, we ensure that only the necessary changes are integrated, reducing the risk of breaking existing functionality.
This feature would significantly improve developer trust and control over AI-generated changes, leading to more efficient collaboration between users and the AI system.
Possible Implementation Ideas:
A dedicated UI for reviewing and approving changes similar to Cline.
A "lock" feature that prevents specific files from being altered by the AI.
Integration with version control systems to provide an additional layer of safety before applying changes.
Beta Was this translation helpful? Give feedback.
All reactions