Conversation
The Claude CLI hides ExitPlanMode, EnterPlanMode and AskUserQuestion from non-interactive sessions that have no permission prompt tool, so the AI panel's plan mode could never propose a plan: the model wrote the plan file and stopped. Pass a canUseTool handler so the CLI runs with --permission-prompt-tool and route each request to a panel card. - ExitPlanMode shows the plan card; approve continues in the same turn (PostToolUse hint says how), revise denies with the feedback so the model re-plans in place. Old stream-pause / abort / re-prompt flow removed. - Other permission asks (Bash or MCP tools in plan mode, writes outside the project) show an Allow/Deny card instead of being denied silently. - Pending browser answers keyed by confirm id so several cards can be up at once; previously only the last card's click did anything. - Tool ids captured at block start so a subagent tool arriving mid-stream no longer steals the parent tool's details and leaves its card spinning. - User follow-ups are read by the main agent only; subagents no longer consume the queue. Background tasks disabled: an auto-backgrounded subagent's tools were cancelled with a message the model read as a user rejection. - Read-only editor MCP tools carry readOnlyHint; SendMessage/TaskOutput/ TaskStop allowed; plan approval lands in Auto mode. - Fullscreen plan overlay styles: pinned footer, Stop/Approve, always-on revise box, neutralised Bootstrap skins so light theme matches dark.
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The Claude CLI hides ExitPlanMode, EnterPlanMode and AskUserQuestion from non-interactive sessions that have no permission prompt tool, so the AI panel's plan mode could never propose a plan: the model wrote the plan file and stopped. Pass a canUseTool handler so the CLI runs with --permission-prompt-tool and route each request to a panel card.