Skip to content

feat(cli,desktop): /plan enters plan mode directly - #289

Merged
oratis merged 1 commit into
mainfrom
claude/plan-command
Sep 3, 2026
Merged

feat(cli,desktop): /plan enters plan mode directly#289
oratis merged 1 commit into
mainfrom
claude/plan-command

Conversation

@oratis

@oratis oratis commented Sep 3, 2026

Copy link
Copy Markdown
Owner

Closes the /plan row in docs/BEHAVIOR_PARITY.md (was 🔄 "set via /mode plan").

What

  • CLI: /plan switches to plan mode, /plan off back to default. Already in the target mode → says so and does nothing.
  • Desktop palette: same command, same semantics, routed through the existing set-mode action.
  • A trailing prompt (/plan refactor auth) is refused with a pointer rather than silently dropped — mode unchanged, so nothing half-happens.
  • Parity row flipped to ✅; /mode plan keeps working.

Verification

typecheck · lint · format:check · docs:check green; CLI 251 passed (+2: enter/leave, trailing-prompt refusal), desktop 105 passed (+1: palette parse cases).

🤖 Generated with Claude Code

/mode plan already did this; /plan is the word Claude Code users type, and
the parity table carried it as a deferred row for exactly that reason.
Registered in both surfaces that own a slash registry — the CLI and the
desktop palette — with the same three behaviours: bare /plan switches to
plan mode, /plan off switches back to default, and a trailing prompt is
refused with a pointer instead of silently dropped, because half-obeying
a command misteaches what it did.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
@oratis
oratis force-pushed the claude/plan-command branch from 1aad6d5 to 9c8bd3d Compare September 3, 2026 01:38
@oratis
oratis merged commit cedeb85 into main Sep 3, 2026
5 checks passed
@oratis
oratis deleted the claude/plan-command branch September 3, 2026 01:42
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant